<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Artificial Intelligence | Salesforce | RPA on Medium]]></title>
        <description><![CDATA[Stories by Artificial Intelligence | Salesforce | RPA on Medium]]></description>
        <link>https://medium.com/@priyankaattari.pa?source=rss-9d37d545daa8------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*HJCPfd3kpD7XzVZz.jpg</url>
            <title>Stories by Artificial Intelligence | Salesforce | RPA on Medium</title>
            <link>https://medium.com/@priyankaattari.pa?source=rss-9d37d545daa8------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 10:29:17 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@priyankaattari.pa/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Artificial Intelligence]]></title>
            <link>https://medium.com/@priyankaattari.pa/artificial-intelligence-a542e58c99db?source=rss-9d37d545daa8------2</link>
            <guid isPermaLink="false">https://medium.com/p/a542e58c99db</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Artificial Intelligence | Salesforce | RPA]]></dc:creator>
            <pubDate>Tue, 12 Nov 2019 10:54:36 GMT</pubDate>
            <atom:updated>2019-11-12T10:54:36.378Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>What is Artificial Intelligence?</strong></p><p>Artificial intelligence (AI) is a branch of computer science that emphasizes the creation of intelligent machines that work and react like humans. With applications varying from SIRI to self-driving cars, artificial intelligence (AI) has been progressing rapidly. While science fiction often portrays AI as robots with human-like characteristics, AI can encompass anything from Google’s search algorithms to IBM’s Watson to autonomous weapons.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/545/1*Z6nSrGhcBCAy90mKK8hWEQ.png" /></figure><h3>Artificial Intelligence vs Machine Learning vs Deep Learning</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/604/1*1-cFD1xhQO0faKleqWD97g.png" /></figure><h3>Types of Artificial Intelligence</h3><p>Artificial Intelligence has a broad range of applications. These can be categorized into four types:</p><p>· <strong>Type 1: Reactive machines</strong></p><p>An IBM chess program called Deep Blue beat Garry Kasparov in the 1990s. It identifies pieces on the chessboard and makes predictions. It analyzes possible moves, its own and its opponent’s and decides what the most strategic move would be. However, it has no memory because of which it cannot use past experiences to inform future ones.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/280/1*Qrf043pMZ396Q5squ7si2g.png" /></figure><p>· <strong>Type 2: Limited memory</strong></p><p>These AI systems can use past experiences to inform future decisions. Some of the decision-making functions in self-driving cars are designed this way.</p><p>· <strong>Type 3: Theory of mind</strong></p><p>This psychology term refers to the understanding that others have their own beliefs, desires, and intentions that impact the decisions they make.</p><p>· <strong>Type 4:</strong> <strong>Self-awareness</strong></p><p>In this category, AI systems have a sense of self, have consciousness. Machines with self-awareness understand their current state and can use the information to infer what others are feeling. This type of AI does not yet exist</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/481/1*7hPbJk1j5QpeRqyJaNgKsA.png" /></figure><h3>Applications of Artificial Intelligence</h3><p>·</p><p><strong>AI in healthcare</strong>. AI plays a very important role in the healthcare sector. From reducing costs to early detection of diseases like cancer, AI is used quite widely in the healthcare industry. Companies are applying machine learning to make better and faster diagnoses than humans. One of the best-known healthcare technologies is IBM Watson. It understands natural language and is capable of responding to questions asked of it. Other AI applications include chatbots, a computer program used online to answer questions and assist customers, to help schedule follow-up appointments or aid patients through the billing process, and virtual health assistants that provide basic medical feedback.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/408/1*8XwQAfsWj5p-JNU7c-dBcg.png" /></figure><p><strong>AI in business</strong>. Robotic process automation is being applied to highly repetitive tasks normally performed by humans. Machine learning algorithms are being integrated into analytics and CRM platforms to uncover information on how to better serve customers. Chatbots have</p><p>· been incorporated into websites to provide immediate service to customers.</p><p>· <strong>AI in education</strong>. Grade automation is arguably the most useful feature for an educator and AI makes this possible. It can assess students and adapt to their needs, enabling them to work at their own pace. AI tutors can provide additional support to students, ensuring they stay on track.</p><p>· <strong>AI in finance</strong>. AI in personal finance applications, such as Mint or Turbo Tax, is disrupting financial institutions. Applications such as these collect personal data and provide financial advice. Other programs, such as IBM Watson, have been applied to the process of buying a home. Today, software performs much of the trading on Wall Street.</p><p>· <strong>AI in law</strong>. The discovery process, sifting through of documents, in law is often overwhelming for humans. Automating this process is a more efficient use of time. Startups are also building question-and-answer computer assistants that can sift programmed-to-answer questions by examining the taxonomy and ontology associated with a database.</p><p>· <strong>AI in manufacturing</strong>. This is an area that has been at the forefront of incorporating robots into the workflow. Industrial robots used to perform single tasks and were separated from human workers, but as technology advanced that changed.</p><h3>Components of Artificial Intelligence</h3><h3>Machine Learning | Learning from experience</h3><p>Machine learning, or ML, is an application of AI that provides computer systems with the ability to automatically learn and improve from experience without being explicitly programmed. ML focuses on the development of algorithms that can analyze data and make predictions. Beyond being used to predict what Netflix movies you might like, or the best route for your Uber, machine learning is being applied to healthcare, pharma, and life sciences industries to aid disease diagnosis, medical image interpretation, and accelerate drug development</p><p><strong>Deep Learning | Self-educating machines</strong></p><p>Deep learning is a subset of machine learning that employs artificial neural networks that learn by processing data. Artificial neural networks mimic the biological neural networks in the human brain. Multiple layers of artificial neural networks work together to determine a single output from many inputs, for example, identifying the image of a face from a mosaic of tiles. The machines learn through positive and negative reinforcement of the tasks they carry out, which requires constant processing and reinforcement to progress.</p><h3>Neural Network | Making associations</h3><p>Neural networks enable deep learning. As mentioned, neural networks are computer systems modeled after neural connections in the human brain. The artificial equivalent of a human neuron is a perceptron. Just like bundles of neurons create neural networks in the brain, stacks of perceptrons create artificial neural networks in computer systems.</p><h3>Cognitive Computing | Making inferences from context</h3><p>Cognitive computing is another essential component of AI. Its purpose is to imitate and improve the interaction between humans and machines. Cognitive computing seeks to recreate the human thought process in a computer model, in this case, by understanding human language and the meaning of images.</p><p>Together, cognitive computing and artificial intelligence strive to endow machines with human-like behaviors and information processing abilities.</p><h3>Natural Language Processing (NLP) | Understanding the language</h3><p>Natural Language Processing or NLP, allows computers to interpret, recognize, and produce human language and speech. The ultimate goal of NLP is to enable seamless interaction with the machines we use every day by teaching systems to understand human language in context and produce logical responses.</p><p>Real-world examples of NLP include Skype Translator, which interprets the speech of multiple languages in real-time to facilitate communication.</p><h3>Computer Vision | Understanding images</h3><p>Computer vision is a technique that implements deep learning and pattern identification to interpret the content of an image; including the graphs, tables, and pictures within PDF documents, as well as, other text and video. Computer vision is an integral field of AI, enabling computers to identify, process and interpret visual data.</p><p>Applications of this technology have already begun to revolutionize industries like research &amp; development and healthcare. Computer Vision is being used to diagnose patients faster by using Computer Vision and machine learning to evaluate patients’ x-ray scans.</p><h3>Machine learning &amp; OpenCV</h3><p>In this part of the AI tutorial you will learn about Machine Learning and OpenCV, Machine Learning workflow, the importance of saving the model, the importance of saving the model, OpenCV, Generative Adversarial Network and more.</p><h3>Why save the model?</h3><p>In this tutorial, we are going to see about the machine learning flow from development to release phase, what is the need of saving a model and basics of OpenCV, GAN.</p><p>We need to save the models whenever we run as a pickle file. Machine learning workflow is these following steps –<br><strong>Understanding Problem statement</strong><br>Gathering the data<br>Data cleaning or pre-processing<br>Divide the data into Train, validation, and Test.<br>Build the model<br>Validate the model &amp; test the model on real data<br>Ensure the Accuracy, True positive and false positive<br>Deploy the model.<br><strong>Become a Master of Artificial Intelligence by going through this online </strong><a href="https://intellipaat.com/artificial-intelligence-deep-learning-course-with-tensorflow/"><strong>Artificial Intelligence training</strong></a><strong>.</strong></p><p>Eg. Image classifier<br>The data (training text, document, images etc.) goes to feature vectors and then processed to the machine learning algorithm. The labels are passed directly to the machine learning algorithm. The typical steps involve training the model, validating and finalizing the model, saving the model and load the saved model for the next prediction. This is a recurring process. We can use either forward or backward propagation to train the model.</p><p><strong>Why should we save the model?</strong><br>The classification should be performed in real-time and need to be deployed in the production server. The challenges we might face are –</p><p>Productions and local environment might not be the same<br>The run-time delay might occur.Classification needs to be done on real-time on a production server. Next time whenever the image comes, the machine has to identify the same in case it matches the previous model already run in the system. From building a model to release it we must might the flow.<br>Prepare yourself for the Top Artificial Intelligence Interview Questions And Answers Now!</p><p><strong>How to save the model?</strong><br><strong>Pickle file:</strong> Pickling, also known as Serializing is a method of converting any python object into a stream of bytes by considering the hierarchy of the object we are trying to convert. The Unpickling, also known as Marshaling is the reverse process of pickling where we convert the stream of bytes to the original python object by considering the same hierarchy which we followed previously.<br><strong>H5 file:</strong> It is a data file saved in the hierarchical data format (HDF). It contains a multidimensional array of scientific data. They are most commonly used in Aerospace, Engineering, Finance, Physics, etc.</p><p><strong>Open CV :</strong><br>OpenCV(Open Source Computer Vision) is an image processing library created by Intel and maintained by Willow Garage. OpenCV was released under BSD (Berkeley Source Distribution) license-free for both academic and commercial. It includes hundreds of computer vision algorithms. Available for C, C++ &amp; python. Newest update OpenCV 3.4 (Dec 2017). It is easy to use and install. These are described in more detail on AI and Deep Learning Community.</p><p><strong>GAN :</strong><br>GANs or Generative Adversarial Network is the class of AI algorithms used in unsupervised machine learning. They are used for image enhancement using automated texture synthesis in combination with perceptual loss. The components of GAN are –</p><ol><li>Generator</li></ol><p>2.Discriminator</p><p>The major applications of GAN are that from the downsampled images i.e. the one with fewer details, the generators tires to approximate them to a more natural detailed version.</p><h3>Back Propagation Algorithm</h3><p>Once the forward propagation is done and the neural network gives out a result, how do you know if the result predicted is accurate enough? This is where the backpropagation algorithm is used to go back and update the weights so that the actual values and predicted values are close enough.</p><p>In an <a href="https://intellipaat.com/blog/what-is-artificial-neural-network/">artificial neural network</a>, the values of weights and biases are randomly initialized. Due to random initialization, the <a href="https://intellipaat.com/blog/tutorial/machine-learning-tutorial/neural-network-tutorial/">neural network</a> probably has errors in giving the correct output. We need to reduce error values as much as possible. So, for reducing these error values, we need a mechanism that can compare the desired output of the neural network with the network’s output that consists of errors and adjusts its weights and biases such that it gets closer to the desired output after each iteration. For this, we train the network such that it back propagates and updates the weights and biases. This is the concept of the backpropagation algorithm.</p><h3>Understanding Deep Learning</h3><ol><li>Parameter Initialization</li><li>Feedforward Propagation</li><li>Backpropagation</li></ol><p>We will look into all these steps, but mainly we will focus on back propagation algorithm.</p><p>Parameter Initialization: In this, parameters, i.e., weights and biases, associated with an artificial neuron are randomly initialized. After receiving the input, the network feed forwards the input and it makes associations with weights and biases to give the output. The output associated to those random values is most probably not correct. So, next, we will see feedforward propagation.</p><p>Want to become master in Artificial Intelligence, check out this Artificial Intelligence Training!</p><p>Feedforward propagation: After initialization, when the input is given to the input layer, it propagates the input into hidden units at each layer. The nodes here do their job without being aware whether results produced are accurate or not (i.e., they don’t re-adjust according to the results produced). Then, finally, the output is produced at the output layer. This is called feedforward propagation.</p><p>Backpropagation in Neural Networks: The principle behind backpropagation algorithm is to reduce the error values in randomly allocated weights and biases such that it produces the correct output. The system is trained in the supervised learning method, where the error between the system’s output and a known expected output is presented to the system and used to modify its internal state. We need to update the weights such that we get the global loss minimum. This is how backpropagation in neural networks works.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/592/1*PO4uIdORv5v2lMrS6ICD2A.png" /></figure><p>When the gradient is negative, an increase in weight decreases the error.</p><p>When the gradient is positive, decrease in weight decreases the error.</p><h3>Working of Back Propagation Algorithm</h3><p>How does back propagation algorithm work?</p><p>The goal of the back propagation algorithm is to optimize the weights so that the neural network can learn how to correctly map arbitrary inputs to outputs. Here, we will understand the complete scenario of backpropagation in neural networks with the help of a single training set.</p><p>n order to have some numbers to work with, here are initial weights, biases, and training input and output.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/625/1*DnZziAg2p2X5K6qUU00XOg.jpeg" /></figure><p><em>he sigmoid function pumps the values for which it is used in the range, 0 to 1.</em></p><p><em>It is used for models where we have to predict the probability. Since the probability of any event lies between 0 and 1, the sigmoid function is the right choice.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/651/0*PCP8D_Qlqs5aaI57.png" /></figure><p>Carrying out the same process for h2:</p><pre>out h2 = 0.596884378</pre><p>The output for o1:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/667/0*IYv4JEfgCAjict3A.png" /></figure><p><em>Carrying out the same process for o2:</em></p><pre>out o2 = 0.772928465</pre><h3>Calculating the Total error:</h3><p>We can now calculate the error for each output neuron using the squared error function and sum them up to get the total error: E total = Ʃ1/2(target — output)2</p><p>The target output for o1 is 0.01, but the neural network output is 0.75136507; therefore, its error is:</p><pre>E o1 = 1/2(target o1 - out o1)2 = 1/2(0.01 - 0.75136507)2 = 0.27481108 ……………..……………. (Equation 5)</pre><p>By repeating this process for o2 (remembering that the target is 0.99), we get:</p><pre>E o2 = 0.023560026</pre><p>Then, the total error for the neural network is the sum of these errors:</p><pre>E total = E o1 + E o2 = 0.274811083 + 0.023560026 = 0.298371109</pre><h3>Step 2: Backward Propagation:</h3><p>Our goal with backpropagation algorithm is to update each weight in the network so that the actual output is closer to the target output, thereby minimizing the error for each output neuron and the network as a whole.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/631/1*3S-WAU3kNpxP7qn2LpXAmA.png" /></figure><p>Consider w5; we will calculate the rate of change of error w.r.t the change in the weight w5:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/615/0*cZA5C88Q9KHwhg9D.png" /></figure><p>Since we are propagating backward, the first thing we need to do is to calculate the change in total errors w.r.t the outputs o1 and o2:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/643/0*5YT8VYsx-Y62FvOk.png" /></figure><p>Now, we will propagate further backward and calculate the change in the output o1 w.r.t to its total net input:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/701/0*XVcqLvz9wm8dIPNv.png" /></figure><p>How much does the total net input of o1 change w.r.t w5?</p><p><strong>Putting all values together and calculating the updated weight value:</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/687/0*ac7PlIlgyzFlVKAF.png" /></figure><p>Let’s calculate the updated value of w5.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/704/0*WTszbQMFRsAtqtRh.png" /></figure><p>We can repeat this process to get the new weights w6, w7, and w8.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/161/0*3MGS5zHGjNiY8IZ-.png" /></figure><p>We perform the actual updates in the neural network after we have the new weights leading into the hidden layer neurons.</p><p><strong>We’ll continue the backward pass by calculating new values for w1, w2, w3, and w4:</strong></p><p>Starting with w1:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/571/0*4yxm6H5k4rcUc2he.png" /></figure><p>We’re going to use a similar process as we did for the output layer, but slightly different to account for the fact that the output of each hidden layer neuron contributes to the final output. Thus, we need to take Eo1 and Eo2 into consideration.</p><p>We can visualize it as below:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/520/1*VO7_-MFSKwNSHx_mDT_y3Q.png" /></figure><p><strong><em>Starting with h1:</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/542/0*rV_W055Lb22RIjOR.png" /></figure><p>We can calculate:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/682/0*pDG5dXP-wmIA9D2M.png" /></figure><p>We will calculate the partial derivative of the total net input of h1 w.r.t w1 the same way as we did for the output neuron.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/227/0*q7mHlUyQwzitMqku.png" /></figure><p>Let’s put it all together.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/564/0*yPIudU9Guvxv0UXP.png" /></figure><ul><li><em>When we fed forward 0.05 and 0.1 inputs originally, the error on the network was 0.298371109.</em></li><li><em>After the first round of backpropagation, the total error is now down to 0.291027924.</em></li></ul><p>It might not seem like much, but after repeating this process 10,000 times, for example, the error plummets to 0.0000351085. At this point, when we feedforward 0.05 and 0.1, the two output neurons will generate 0.015912196 (vs. 0.01 target) and 0.984065734 (vs. 0.99 target).</p><p>Now, in this back propagation algorithm blog, let’s go ahead and comprehensively understand “Gradient Descent” optimization.</p><p><strong><em>Prepare yourself for the </em></strong><a href="https://intellipaat.com/interview-question/artificial-intelligence-interview-questions/"><strong><em>Artificial Intelligence Interview questions and answers</em></strong></a><strong><em> Now!</em></strong></p><h3>Understanding Gradient Descent</h3><ul><li>Gradient descent is by far the most popular optimization strategy used in <a href="https://intellipaat.com/blog/tutorial/machine-learning-tutorial/">Machine Learning</a> and <a href="https://intellipaat.com/blog/tutorial/machine-learning-tutorial/introduction-deep-learning/">Deep Learning</a> at the moment. It is used while training our model, can be combined with every algorithm, and is easy to understand and implement.</li><li>Gradient measures how much the output of a function changes if we change the inputs a little.</li><li>We can also think of a gradient as the slope of a function. The higher the gradient, the steeper the slope and the faster the model learns.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/140/0*OcTeAeftjDeClduW.png" /></figure><p>where,</p><p><em>b</em> = next value</p><p><em>a</em> = current value</p><p>‘−’ refers to the minimization part of the gradient descent.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/693/0*RCJVHwncQ2z5ZJ0V.png" /></figure><ul><li>This formula basically tells us the next position where we need to go, which is the direction of the steepest descent.</li><li>Gradient descent can be thought of as climbing down to the bottom of a valley, instead of as climbing up a hill. This is because it is a minimization algorithm that minimizes a given function.</li><li>Let’s consider the graph below where we need to find the values of w and b that correspond to the<em> minimum cost function </em>(marked with a red arrow).</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/572/1*HlT_X5ls8TMvGnHB1YCQsQ.png" /></figure><ul><li>To start with finding the right values, we initialize the values of <em>w</em> and <em>b</em> with some random numbers, and gradient descent starts at that point (somewhere around the top).</li><li>Then, it takes one step after the other in the steepest downside direction (e.g., from top to bottom) till it reaches the point where the cost function is as small as possible.</li></ul><p><strong>Batches</strong></p><ul><li>The total number of training examples present in a single batch is referred to as the batch size.</li><li>Since we can’t pass the entire dataset into the neural net at once, we divide the dataset into the number of batches or sets or parts.</li></ul><p>Moving ahead in this blog on the “Back Propagation Algorithm”, we will look at the types of gradient descent.</p><h3>Types of Gradient Descent</h3><p><strong>Batch Gradient Descent</strong></p><p>In batch gradient descent, we use the complete dataset available to compute the gradient of the cost function. Batch gradient descent is very slow because we need to calculate the gradient on the complete dataset to perform just one update, and if the dataset is large then it will be a difficult task.</p><ol><li>Cost function is calculated after the initialization of parameters.</li><li>It reads all the records into memory from the disk.</li><li>After calculating sigma for one iteration, we move one step further and repeat the process.</li></ol><p><strong>Mini-batch Gradient Descent</strong></p><p>It is a widely used algorithm that makes faster and accurate results. The dataset, here, is clustered into small groups of ’n’ training datasets. It is faster because it does not use the complete dataset. In every iteration, we use a batch of ’n’ training datasets to compute the gradient of the cost function. It reduces the variance of the parameter updates, which can lead to more stable convergence. It can also make use of a highly optimized matrix that makes computing of the gradient very efficient.</p><p><strong>Stochastic Gradient Descent</strong></p><p>We use stochastic gradient descent for faster computation. The first step is to randomize the complete dataset. Then, we use only one training example in every iteration to calculate the gradient of the cost function for updating every parameter. It is faster for larger datasets also because it uses only one training example in each iteration.</p><p>We understood all the basic concepts and working of backpropagation algorithm through this blog. Now, we know that backpropagation algorithm is the heart of a neural network.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a542e58c99db" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What is Artificial Intelligence? A Beginners Guide to Artificial Intelligence]]></title>
            <link>https://medium.com/@priyankaattari.pa/what-is-artificial-intelligence-a-beginners-guide-to-artificial-intelligence-f212e3ff79f5?source=rss-9d37d545daa8------2</link>
            <guid isPermaLink="false">https://medium.com/p/f212e3ff79f5</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Artificial Intelligence | Salesforce | RPA]]></dc:creator>
            <pubDate>Mon, 11 Nov 2019 13:49:01 GMT</pubDate>
            <atom:updated>2019-11-11T13:49:01.205Z</atom:updated>
            <content:encoded><![CDATA[<p>Take a minute. Think about what you are doing. Imagine yourself 10 years from now. Where do you think the world is heading? Have you ever wondered how fast the world is advancing? Have you ever thought about how revolutionary the internet was?</p><p>Everything we do revolves around this thing called the Internet. Now that I think about it, without the internet, everything would become so obsolete. It’s like living in another time era, we have gone from historic to modern within a few decades; a blink of an eye on the earth’s timeline of existence.</p><p>I think it’s less than 10 years. And It lies within this term: <strong>Artificial Intelligence.</strong></p><p>It’s as simple as:</p><p><strong>Artificial + Intelligence</strong></p><p><strong>Artificial</strong> means something that’s not natural</p><p><strong>Intelligence</strong> the ability to acquire and apply knowledge and skills</p><p>AI = Crafted, intelligent behavior by machines.</p><p>If we delve deeper, AI can be divided into two categories:</p><p><strong>General AI: </strong>Would have all of the characteristics of human intelligence</p><p><strong>Narrow AI: </strong>Exhibits some facets of human intelligence, sort of like specialization.</p><p>· A machine that’s great at recognizing images, but nothing else is an example of Narrow AI</p><p>Easy, right? And there’s so much more.</p><p>It’s amazing how AI is around you. Now, your first view towards AI would be walking talking robots but it doesn’t have to be that way. Look at the image above. That’s Artificial intelligence!</p><p>Deep learning — — — Machine Learning — — — — Artificial Intelligence</p><p>That’s the basic flow between the terms. Let’s start with <strong>Machine Learning</strong>. Also to know an important fact Machine Learning is a subset of Artificial Intelligence.</p><p><strong>Machine Learning:</strong></p><p>· Works by combining large amounts of data + fast, iterative processing and intelligent algorithms allowing the software to learn automatically from patterns or features in data.</p><p>At its core, Machine Learning is a method achieving AI; it learns from example. AI is possible without Machine Learning but would require billions of lines of code.</p><p>A Machine without data is like a newborn baby. As the baby grows, it learns through the countless images taken through the eyes, through language and through experience which is generally a form of trial and error.</p><p>As the saying goes, you learn from your mistakes. Even Machines do. The <strong>key takeaway</strong> here is that humans and machines <strong>learn the same way</strong>! The only difference is that machines are much <strong>faster</strong> and more <strong>efficient</strong>.</p><p>Visual representation of the <strong>branches of Machine Learning</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/1*9XSICB1QnVZ28F7vkyHklg.png" /></figure><p><strong>But where can we use Machine Learning?</strong></p><p>Machine Learning allows for the automation of many things.</p><p>Let’s take a use case: <strong>Self-Driving Cars</strong></p><p>Self-Driving cars use Machine Learning to “sense” the environments. This is achieved through the sensors and software integrated into the car which conduct real-time analysis of its surroundings. The sensors take pictures, record sounds and feed it to the Machine which then makes decisions based on the data. And here we have a car which can drive on its own. Long drives just become a lot less tiring!</p><p><strong>Deep Learning:</strong></p><p>Deep Learning is a method of achieving Machine Learning. It is by far the best method of achieving Machine Learning and its quite an interesting concept.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/606/1*2YdrveRjDuOuRXmCfWni8w.png" /></figure><p>The <strong>human brain</strong> is the most advanced cognitive “machine” in our human body. It works as a network of neurons are fired together. Deep Learning mimics this same process.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/587/1*NLhGwTTzhm7XdzrlrXHxJw.png" /></figure><p>A <strong>neuron</strong> contains of <strong>three</strong> main parts: a <strong>cell body</strong> that directs all activities of the neurons, <strong>dendrites</strong> (tree branches) which are short fibres that receive information from other neurons and pass those messages to the cell body and finally, the <strong>axon</strong> a long single fibre that transmits messages from the <strong>cell body to dendrites of other neurons.</strong></p><p>A machine neuron is known as the <strong>perceptron</strong>. The perceptrons are interconnected through binary networks that work together to create Artificial Neural Networks (ANN’s). The perceptrons function like neurons, they take inputs from different sources and output binary digits (0,1)</p><p>By connecting many perceptrons we create a neural network layer.</p><p>A network of layers creates a neural network. This is termed Deep Learning since the network is “multiple layers deep”</p><p>This is an example of an <strong>Artificial Neural Network</strong>. It consists of multiple layers.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/606/1*YQSXKwonX_989z8mharo6g.png" /></figure><p>In this network:</p><p>1. <strong>The Input Layer: </strong>Here the data is fed to the machine. This includes in this example, the portraits of many people.</p><p>2. <strong>Hidden Layer 1:</strong> It learns to identify contours or basic facial elements</p><p>3. <strong>Hidden Layer 2: </strong>Based on contours and initial analysis, the machine creates more complex facial features such as eyes, nose, mouth and starts to identify the complex facial elements</p><p>4. <strong>Hidden Layer 3:</strong> The machine pieces together the face and compares it with the data that was fed. It tries to identify recurrent patterns and features from previous examples</p><p>5. <strong>Output Layer: </strong>Provides the answer based upon analysis of Data, in this case, the image of an entire face.</p><p>This is overly simplified and a very basic representation of Deep Learning. It can get much more complex. But along with complexity, it brings so much more potential!</p><p><strong>The key takeaway</strong> here is that Deep Learning<strong> mimics</strong> the neurons and neural systems in the human body in order to <strong>achieve</strong> Machine Learning.</p><p>Like Deep Learning, there are many more methods of achieving Machine Learning. These include <strong>Bayesian networks</strong>, <strong>support vector machine (SVM), boosted and bagged decision trees, <em>k</em>-nearest neighbor, Naïve Bayes, discriminant analysis, and logistic regression.</strong></p><p>Artificial Intelligence is a vast field. The more you dig deeper, the more you there’s potential. Artificial Intelligence will, in my opinion, become the core backbone of many technologies to come in the future. AI is useful in any industry including healthcare, banking, manufacturing, IT and just about anything!</p><p>Artificial Intelligence is a vast field. The more you dig deeper, the more you there’s potential. Artificial Intelligence will, in my opinion, become the core backbone of many technologies to come in the future. AI is useful in any industry including healthcare, banking, manufacturing, IT and just about anything!</p><h3>AI vs ML vs DL</h3><p>Human beings are at a new era of technology, and this era is influenced by Artificial Intelligence, Machine Learning, and Deep Learning. It is true that people are mixing up these three terms. However, these terms lie in a hierarchy and can have different applications. So, in this blog, we will look at Artificial Intelligence vs Machine Learning and Deep Learning vs Machine Learning, compare all the three terms as Artificial Intelligence vs Machine Learning vs Deep Learning, and understand these terms in detail.</p><p>Understanding Artificial Intelligence vs Machine Learning vs Deep Learning<br>We know that Earth is surrounded by an atmosphere, and it comprises layers of the atmosphere. The layer which is suitable for human beings to survive is the troposphere. There are three more layers which we would not usually discuss about. However, the key focus is that the atmosphere is the umbrella term for all the layers that reside in it. Similarly, Artificial Intelligence is the umbrella term, or we can say that it is the presentation layer under which Machine Learning and Deep Learning exists.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/1*my8MzaY7Omb6stHgeE01xw.jpeg" /></figure><h3>Artificial Intelligence</h3><p>According to John McCarthy, ‘The science and engineering of making intelligent machines, especially intelligent computer programs’ is Artificial Intelligence.</p><p>There are two categories of Artificial Intelligence:</p><p><strong>Weak Artificial Intelligence</strong>: In this, the machine programs act according to a well-defined response. They are confined to a set of rules that we provide, and they give a response within the domain of those rules.</p><p>A common example can be a home appliance like an oven. We can set the timer and temperature in it according to the need. It performs the task according to the given instruction. It does not have the ability to make decisions and make any changes by itself. So, it comes under the category of Weak Artificial Intelligence.</p><p><strong>Strong Artificial Intelligence: </strong>Machine Learning and Deep Learning comes under the category of Strong Artificial Intelligence. It involves designing of algorithms for machines that try to learn by themselves using the input data and improve the accuracy in giving outputs.</p><p>Examples of Strong Artificial Intelligence are speech recognition, visual perception, and language translation. In language translation, a machine extracts the meaning of words and then the meaning of sentences. After that, it searches for similar-meaning words and sentences in another language and then translates them to that language. As it requires many tasks to accomplish translation and involves decision-making to put everything right, it comes under the category of Strong Artificial Intelligence.</p><p>By now, we have seen ‘What is Artificial Intelligence?’ After this, to make the picture more clear for Artificial Intelligence vs Machine Learning vs Deep Learning, we will now move on to Machine Learning.</p><p><strong>Machine Learning</strong><br>Machine Learning is the subset of AI where Machine Learning algorithms are designed in such a way that the machine tries to learn by itself without being explicitly programmed on each and every instruction. So, as it will be exposed to more and more data, it tries to internally modify itself and adjust according to the data to which it is exposed so that it will not rely on human experts to program them.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/1*tmRLsNK7U8AvEzCOXPCx6A.jpeg" /></figure><p><strong>Supervised Learning</strong><br>In supervised learning, the machine is provided with the labeled dataset. It already has input and output parameters. So, when the machine is given a new dataset, the supervised learning algorithm examines the data and produces the correct output according to the labeled data.</p><p>Use case: Detecting cancer patients</p><p>There would be some parameters and symptoms given for detecting cancer patients. The machine will try to classify the patients according to the symptoms to determine whether they are having cancer or not. So, supervised learning is best for classification and regression problems.</p><p><strong>Unsupervised Learning</strong></p><p>In unsupervised learning, the machine would not have any labeled dataset. The algorithm is designed in a way that tries to learn by itself without any supervision of data. This involves the clustering of data.</p><p>Example: Consider a few objects such as a pencil, eraser, and matchbox</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/630/1*5lGOzG4UVHXo03pWqCu-eA.png" /></figure><p>Here, the machine does not even know what these objects are; rather, it makes clusters of similar objects, and when any input dataset is given, it gives the output by examining the data it has clustered.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/616/1*C1fnynqARufgcAKkMgX5Ag.png" /></figure><p>Netflix&#39;s recommendation system works on the same technique as it saves the users’ watched history and recommends similar content to the user.</p><h3>Reinforcement Learning</h3><p>In reinforcement learning, the algorithms are designed in such a way that the machine tries to find an optimal solution. It adopts the principle of reward and punishment, and by this approach, it moves to the correct result.</p><p>Consider a scenario where a young cricketer tries to learn the technique of hitting a 6 for a particular shot. Whenever he tries to play the shot and misses it, he gets a score of −1.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/304/1*poxIYFaK1kaGoHj2qQ_Qmw.png" /></figure><p>The bowler bowls again, and the batsman tries to hit that shot by adjusting his position. After trying for five to six times, he finally hits a 6 with a suitable position and gets a score of +6.</p><p>This way he learns the technique for hitting a 6. The reinforcement algorithm works in a similar way.</p><h3>Types of Machine Learning at a Glance</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/630/1*gkK4vANd5hNOfTQJKDWVeQ.png" /></figure><p><strong>Deep Learning</strong><br>Deep Learning has evolved from Machine Learning. It works in a layered architecture and uses the artificial neural network, a concept inspired by the biological neural network. The human brain usually analyzes and converts the information it receives and tries to identify it from the past information the brain has stored. The brain does this through labeling and assigning information into various groups in a fraction of a millisecond.</p><p>In a similar way, Deep Learning algorithms are trained to identify patterns and classify various types of information to give the desired output when it receives an input.</p><p>So, to understand Deep Learning vs Machine Learning, you must know the difference between Machine Learning and Deep Learning and the major difference is that we need to provide the features manually in Machine Learning. But in Deep Learning, it automatically extracts features for classification which in turn demands a huge amount of data for training DL algorithms. So, in Deep Learning, the accuracy of the output depends on the amount of data.</p><p>Now, let us look at artificial neural networks in Deep Learning.</p><h3>Artificial Neural Network (ANN)</h3><p>It is a concept inspired by the biological neural network. It consists of three layers:</p><ul><li><strong>Input Layer</strong>: The input layer is used for taking the input data from external sources and then passing it on to the hidden layers of the neural network. It does not perform any computation.</li><li><strong>Hidden Layer</strong>: This layer consists of many hidden layers. All the computation is performed in this layer. After all the computation is done, it passes the output to the output layer.</li><li><strong>Output Layer</strong>: This layer is used for computing and giving the output to the outside world.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/1*lXQpmPjvJ5YyulqdCtsGUw.jpeg" /></figure><p>These layers consist of nodes that are interconnected with each other. The nodes interact with each other with the help of links by which they are connected. This connection of nodes is designed in such a way that it produces output for a given input.</p><p>The links are associated with a real number that is called the weight of those links. These weights are initialized randomly, and hence there could be a large difference between the actual values and the predicted values. Due to this, it will not give the desired output in one iteration.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/1*ZgitfZ3AtuDs2PZOQxx9CA.jpeg" /></figure><p>Even after the weight is assigned and the computation is done if it does not give the desired output, we go back and <strong>update</strong> <strong>the weight</strong> of the link with the current value to get closer to the desired output. We do this progressively until we get the best possible output.</p><p>Also, the weights assigned to the links decides how fast the triggering of the <strong>activation function</strong> will occur.</p><p>There is another term ‘<strong>bias</strong>,’ which is used to decide when to trigger the activation function.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/1*bSoOPGoTVaYxYwTgae_wiQ.jpeg" /></figure><p>Now, we will see the details of the activation function.</p><p>In the real world, the data is always 3-dimensional. For example, let’s say, an image of a car is given as an input to the neural network; we can plot the length and height of the car in a 2-D plane. However, there are a lot more attributes to be considered while computing to recognize it as a car. The computation of this data is a complex task. So, to reduce this difficulty, we use the <strong>activation function</strong>. It pumps the wide range of values assigned to the data into a specific domain so that computation becomes easy.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/1*fIk6jf2Wi7jtVl7-80MhXQ.jpeg" /></figure><p>This is how the artificial neural network works and helps in achieving perfection in Artificial Intelligence.</p><p>To summarize, Artificial Intelligence is an umbrella term, and Machine Learning and Deep Learning are the subdomains of this field that help in achieving Artificial Intelligence.</p><p>Hopefully, this tutorial gave the hierarchical description of Artificial Intelligence, Machine Learning, and Deep Learning and cleared the confusion among these terms.</p><p>This is all about Artificial Intelligence vs Machine Learning vs Deep Learning.</p><p>Learn more about Artificial Intelligence from this <a href="https://intellipaat.com/artificial-intelligence-deep-learning-course-with-tensorflow/">Artificial Intelligence Training</a> to get ahead in your career!</p><h3>Artificial Neural Networks</h3><p>Here, in this section of the AI tutorial, you will learn what is an Artificial Neural Network (ANN), detailed architecture of ANN, what is an activation function, forward and backward propagation, importance of Hyperparameter in Artificial Neural Networks.</p><p>Artificial Neural networks<br>ANN was developed considering the same as of our brain, the same as how our brain works were taken into account. It was inspired by the way neurons work, the major task is to process information. The architecture of the neural network is similar to neurons.<br>Frank Rosenblatt in 1958 invented ANN and built the machine learning algorithm.<br>Learn more about Artificial Neural networks in this insightful Artificial Intelligence Training now!</p><h3>Neural Network Components</h3><p>Each neural network consists of –</p><ul><li>Perceptron — a mathematical representation of the neuron</li><li>Weights, Bias — signifies the importance of each parameter.</li><li>Activation function — at each neuron what makes it to be active or inactive</li><li>Outcome</li><li>Backpropagation</li></ul><p>Data can be of any format — Linear and Nonlinear. The neural networks learn the data types based on the activation function. It can understand the data based on quadratic functions.</p><h3>Architecture of Neural network</h3><p>It consists of the input value and output value. Each input value is associated with its weight, which passes on to the next level, each perceptron will have an activation function. The weights and input value forms a single perception. We use activation function and based on that, the value goes to next well. And the process continues till it reaches output y’.<br>Sometimes in nonlinear data, the classification is done in three-dimensionality than in two dimensions.<br><strong>Forward propagation : </strong><br>The factor 1 keeps moving forward and gets activated in the next level, at the nodes the activation value like sigma, return h, based on that, values like 0, 1 or 2 is passed on to the next level.<br><strong>Activation functions :</strong><br><strong> </strong>These are used to activate neurons. Different activation functions are –<br>The activation function produces the output value 0 or 1, i.e. it classifies the output image for the result.</p><p><strong>Cost function :</strong><br>Sometimes the algorithm we create might predict the value incorrectly, so we need cost function. It tried to quantify the error factor of neural network. It calculates how well the neural network is performing based on the actual vs predicted value.<br><strong>Error factor = Predicted — Actual.</strong><br>Cost function is used to minimize the error factor.</p><p>Back Propagation :<br>When we feel that outputs are not correct, we backpropagate the values to adjust the weights to produce the right output. The architecture, activation functions remains the same in each perceptron. Adjusts using gradient descent.</p><p><strong>Hyperparameters of ANN :</strong><br>Hyperparameters are those which are used to tune a neural network. These include –</p><ul><li>Learning rate — how fast it abandons the old belief for new ones</li><li>Momentum — Smooth learning is maintained by gradient descent.</li><li>Epoch — complete forward and backward propagation. As epoch increases, error decreases.</li></ul><p><strong>Also, can it replace humans? There are some disadvantages that come along with AI.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/275/1*ubpNjEx6h7-AzrMz5WPCZQ.jpeg" /></figure><p><strong>Following are some of the disadvantages :</strong><br>1.<strong> High Cost:</strong><br>The creation of artificial intelligence requires huge costs as they are very complex machines. Their repair and maintenance require huge costs.<br>They have software programs that need frequent up-gradation to cater to the needs of the changing environment and the need for the machines to be smarter by the day.<br>In the case of severe breakdowns, the procedure to recover lost codes and reinstating the system might require huge time and cost.<br>2<strong>. No Replicating Humans:</strong><br>Intelligence is believed to be a gift of nature. An ethical argument continues, whether human intelligence is to be replicated or not.<br>Machines do not have any emotions and moral values. They perform what is programmed and cannot make the judgment of right or wrong. Even cannot take decisions if they encounter a situation unfamiliar to them. They either perform incorrectly or breakdown in such situations.<br>3. <strong>No Improvement with Experience:</strong><br>Unlike humans, artificial intelligence cannot be improved with experience. With time, it can lead to wear and tear. It stores a lot of data but the way it can be accessed and used is very different from human intelligence.<br>Machines are unable to alter their responses to changing environments. We are constantly bombarded by the question of whether it is really exciting to replace humans with machines.<br>In the world of artificial intelligence, there is nothing like working with a whole heart or passionately. Care or concerns are not present in the machine intelligence dictionary. There is no sense of belonging or togetherness or a human touch. They fail to distinguish between a hardworking individual and an inefficient individual.<br>4. <strong>No Original Creativity:</strong><br>These are not the forte of artificial intelligence. While they can help you design and create, they are no match to the power of thinking that the human brain has or even the originality of a creative mind.<br>Human beings are highly sensitive and emotional intellectuals. They see, hear, think and feel. Their thoughts are guided by the feelings which completely lacks in machines. The inherent intuitive abilities of the human brain cannot be replicated.<br>5. <strong>Unemployment:</strong><br>Replacement of humans with machines can lead to large scale unemployment.<br>Humans can unnecessarily be highly dependent on the machines if the use of artificial intelligence becomes rampant. They will lose their creative power and will become lazy. Also, if humans start thinking in a destructive way, they can create havoc with these machines.</p><p>To conclude, Some AI experts predict that AI will be able to do anything that humans can but do it better. This is a questionable assumption, but AI will surely surpass humans in specific domains. A chess computer beating the world chess champion was the first example.<br>If AI were to develop to the point that it can do everything better than humans, it would mean that it would also do better in science and technology. It may decide that it is no longer worthwhile to develop a certain field of research — or it may decide space travel is a waste of time as long as humans on earth are living in poverty and more than a billion people have no access to clean drinking water.<br>Most scenarios about future AI are hypothetical, but AI presents us with existential questions that are yet to be answered.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f212e3ff79f5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Artificial intelligence]]></title>
            <link>https://medium.com/@priyankaattari.pa/artificial-intelligence-3594a09abd6?source=rss-9d37d545daa8------2</link>
            <guid isPermaLink="false">https://medium.com/p/3594a09abd6</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[ai-course]]></category>
            <dc:creator><![CDATA[Artificial Intelligence | Salesforce | RPA]]></dc:creator>
            <pubDate>Sat, 09 Nov 2019 11:55:23 GMT</pubDate>
            <atom:updated>2019-11-09T11:55:23.590Z</atom:updated>
            <content:encoded><![CDATA[<p>Artificial intelligence is an area of computer science that emphasizes the creation of intelligent machines, i.e., machines that react and think like humans. As a matter of fact, AI can improve efficiency and productivity by automating repetitive tasks. Additionally, it can create an immersive, responsive experience and understand human emotions. Although it sounds like an easy thing to do, it is quite a tedious task. Machines can often act and react like humans only if they have abundant information relating to the world. Artificial intelligence must have access to objects, categories, properties, and relations between all of them to implement knowledge engineering. Initiating common sense, reasoning and problem-solving power in machines is a difficult and tedious task. And that is why some of the core problems of artificial intelligence include programming computers for certain traits such as: Learn more about Artificial Neural networks in this insightful AI course now!</p><ul><li>Knowledge</li><li>Reasoning</li><li>Problem-solving</li><li>Perception</li><li>Learning</li><li>Planning</li><li>Ability to manipulate and move objects</li></ul><p><strong>Where is Artificial Intelligence used?</strong></p><p>The word artificial intelligence was coined in the very early days of 1956 but it has recently started to take over the world. It is almost impossible for us to not use it. Everyone uses Google Maps, or Alexa or Siri. All these are prime examples of artificial intelligence. Some other applications could be: Check more in-depth about Activation Function from this <a href="https://intellipaat.com/artificial-intelligence-course-bangalore/">artificial Intelligence Course in Bangalore.</a></p><p><strong>AI in marketing:</strong></p><p>Ever wondered how Amazon or Netflix shows you exactly the type of products you want to buy or the movies you want to watch? In a matter of seconds, we get a list of relevant items. It examines millions of records to suggest shows and films that you might like based on your previous actions and choices of films. As the data set grows, this technology is getting smarter and smarter every day. With the growing advancement in AI, in the near future, it may be possible for consumers on the web to buy products by snapping a photo of it. Softwares like CamFind and GoogleLens are already experimenting with this.</p><p><strong>AI in banking:</strong></p><p>A lot of banks have already adopted AI-based systems. A lot of banks have already adopted AI-based systems to provide customer support, detect anomalies and credit card frauds. An example of this is the HDFC Bank.HDFC Bank has developed an AI-based chatbot called EVA (Electronic Virtual Assistant)</p><p><strong>AI in agriculture:</strong></p><p>AI can help farmers get more from the land while using resources more sustainably. Issues such as climate change, population growth, and food security concerns have pushed the industry into seeking more innovative approaches to improve crop yield. Organizations are using automation and robotics to help farmers find more efficient ways to protect their crops from weeds.</p><p><strong>AI in healthcare:</strong></p><p>When it comes to saving our lives, a lot of organizations and medical care centers are relying on AI. There are devices that can detect heart strokes just by analyzing the vitals of a patient and prevent a heart attack. A company called Aifloo is developing a system for keeping track of how people are doing in nursing homes, home care, etc. The best thing about AI in healthcare is that you don’t even need to develop a new medication. Just by using an existing medication in the right way, you can also save lives.</p><p><strong>AI in gaming:</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*M7h62aZpMC1Y5ful2yfZ8A.jpeg" /><figcaption><strong>Image source </strong><a href="https://medium.com/"><strong>Medium.com</strong></a></figcaption></figure><p>And of course, we have to consider the endless scope of using artificial intelligence in the gaming industry. The “play with computer” feature is based on and developed using Artificial Intelligence. Millions of test cases are run through the computer so that it learns the different set of rules and then the computer can play the game too!</p><p><strong>Technologies used in AI:</strong></p><p>1. <strong>Decision Management:</strong> Engines that insert rules and logic into AI systems and used for initial setup/training and ongoing maintenance and tuning. A mature technology, it is used in a wide variety of enterprise applications, assisting in or performing automated decision-making. Sample vendors: Advanced Systems Concepts, Informatica, Maana, Pegasystems, UiPath.</p><p>2. <strong>Natural Language Generation:</strong> Producing text from computer data. Currently used in customer service, report generation, and summarizing business intelligence insights. Sample vendors: Attivio, Automated Insights, Cambridge Semantics, Digital Reasoning, Lucidworks, Narrative Science, SAS, Yseop.</p><p>3. <strong>AI-optimized Hardware: </strong>Graphics processing units (GPU) and appliances specifically designed and architected to efficiently run AI-oriented computational jobs. Currently primarily making a difference in deep learning applications. Sample vendors: Alluviate, Cray, Google, IBM, Intel, Nvidia.</p><p>4. Biometrics: Enable more natural interactions between humans and machines, including but not limited to image and touch recognition, speech, and body language. Currently used primarily in market research. Sample vendors: 3VR, Affectiva, Agnitio, FaceFirst, Sensory, Synqera, Tahzoo.</p><p>5. <strong>Text Analytics and NLP:</strong> Natural language processing (NLP) uses and supports text analytics by facilitating the understanding of sentence structure and meaning, sentiment, and intent through statistical and machine learning methods. Currently used in fraud detection and security, a wide range of automated assistants, and applications for mining unstructured data. Sample vendors: Basis Technology, Coveo, Expert System, Indico, Knime, Lexalytics, Linguamatics, Mindbreeze, Sinequa, Stratifyd, Synapsify.</p><p>6. <strong>Robotic Process Automation:</strong> Using scripts and other methods to automate human action to support efficient business processes. Currently used where it’s too expensive or inefficient for humans to execute a task or a process. Sample vendors: Advanced Systems Concepts, Automation Anywhere, Blue Prism, UiPath, WorkFusion.</p><p>7. <strong>Virtual Agents: </strong>They may vary from simple chatbots to advanced systems that can network with humans. Currently used in customer service and support and as a smart home manager. Sample vendors: Amazon, Apple, Artificial Solutions, Assist AI, Creative Virtual, Google, IBM, IPsoft, Microsoft, Satisfi.</p><p><strong>Now Understand Artificial Intelligence vs Machine Learning vs Deep Learning</strong></p><p>We know that Earth is surrounded by an atmosphere, and it comprises layers of atmosphere. The layer which is suitable for human beings to survive is the troposphere. There are three more layers which we would not usually discuss. However, the key focus is that the atmosphere is the umbrella term for all the layers that reside in it. Similarly, Artificial Intelligence is the umbrella term, or we can say that it is the presentation layer under which Machine Learning and Deep Learning exists.</p><p><strong>To understand Artificial Intelligence vs Machine Learning vs Deep Learning, we will first look at Artificial Intelligence.</strong></p><p><strong>Artificial Intelligence</strong></p><p>According to John McCarthy, ‘The science and engineering of making intelligent machines, especially intelligent computer programs’ is Artificial Intelligence.</p><p>To understand Artificial Intelligence vs Machine Learning, let’s look at the categories of Artificial Intelligence.</p><p><strong>There are two categories of Artificial Intelligence:</strong></p><p><strong>Weak Artificial Intelligence:</strong> In this, the machine programs act according to a well-defined response. They are confined to a set of rules that we provide, and they give a response within the domain of those rules. A common example can be a home appliance like an oven. We can set the timer and temperature in it according to the need. It performs the task according to the given instruction. It does not have the ability to make decisions and make any changes by itself. So, it comes under the category of Weak Artificial Intelligence.</p><p><strong>Strong Artificial Intelligence:</strong> Machine Learning and Deep Learning comes under the category of Strong Artificial Intelligence. It involves designing of algorithms for machines that try to learn by themselves using the input data and improve the accuracy in giving outputs.</p><p>Examples of Strong Artificial Intelligence are speech recognition, visual perception, and language translation. In language translation, a machine extracts the meaning of words and then the meaning of sentences. After that, it searches for similar-meaning words and sentences in another language and then translates them to that language. As it requires many tasks to accomplish translation and involves decision-making to put everything right, it comes under the category of Strong Artificial Intelligence.</p><p><strong>Machine Learning</strong></p><p>Machine Learning is the subset of AI where Machine Learning algorithms are designed in such a way that the machine tries to learn by itself without being explicitly programmed on each and every instruction. So, as it will be exposed to more and more data, it tries to internally modify itself and adjust according to the data to which it is exposed so that it will not rely on human experts to program them.</p><p><strong>Supervised Learning</strong></p><p>In supervised learning, the machine is provided with the labeled dataset. It already has input and output parameters. So, when the machine is given a new dataset, the supervised learning algorithm examines the data and produces the correct output according to the labeled data.</p><p><strong>Use case: Detecting cancer patients</strong></p><p>There would be some parameters and symptoms given for detecting cancer patients. The machine will try to classify the patients according to the symptoms to determine whether they are having cancer or not. So, supervised learning is best for classification and regression problems.</p><p><strong>Unsupervised Learning</strong></p><p>In unsupervised learning, the machine would not have any labeled dataset. The algorithm is designed in a way that it tries to learn by itself without any supervision of data. This involves the clustering of data.</p><p>Example: Consider a few objects such as a pencil, eraser, and matchbox.</p><p>Here, the machine does not even know what these objects are; rather, it makes clusters of similar objects, and when any input dataset is given, it gives the output by examining the data it has clustered.</p><p>Netflix’s recommendation system works on the same technique as it saves the users’ watched history and recommends a similar content to the user.</p><p><strong>Reinforcement Learning</strong></p><p>In reinforcement learning, the algorithms are designed in such a way that the machine tries to find an optimal solution. It adopts the principle of reward and punishment, and by this approach, it moves to the correct result.</p><p>Consider a scenario where a young cricketer tries to learn the technique of hitting a 6 for a particular shot. Whenever he tries to play the shot and misses it, he gets a score of −1.</p><p>The bowler bowls again, and the batsman tries to hit that shot by adjusting his position. After trying for five to six times, he finally hits a 6 with a suitable position and gets a score of +6. This way he learns the technique for hitting a 6. The reinforcement algorithm works in a similar way.</p><p><strong>Deep Learning</strong></p><p>Deep Learning has evolved from Machine Learning. It works in a layered architecture and uses the artificial neural network, a concept inspired by the biological neural network. The human brain usually analyzes and converts the information it receives and tries to identify it from the past information the brain has stored. The brain does this through labeling and assigning information into various groups in a fraction of a millisecond.</p><p>In a similar way, Deep Learning algorithms are trained to identify patterns and classify various types of information to give the desired output when it receives an input.</p><p>So, to understand Deep Learning vs Machine Learning, you must know the difference between Machine Learning and Deep Learning and the major difference is that we need to provide the features manually in Machine Learning. But in Deep Learning, it automatically extracts features for classification which in turn demands a huge amount of data for training DL algorithms. So, in Deep Learning, the accuracy of the output depends on the amount of data.</p><p><strong>Artificial Neural Network (ANN)</strong></p><p>It is a concept inspired by the biological neural network. It consists of three layers:</p><ul><li><strong>Input Layer: </strong>The input layer is used for taking the input data from external sources and then passing it on to the hidden layers of the neural network. It does not perform any computation.</li><li><strong>Hidden Layer:</strong> This layer consists of many hidden layers. All the computation is performed in this layer. After all the computation is done, it passes the output to the output layer.</li><li><strong>Output Layer: </strong>This layer is used for computing and giving the output to the outside world.</li></ul><p>These layers consist of nodes that are interconnected with each other. The nodes interact with each other with the help of links by which they are connected. This connection of nodes is designed in such a way that it produces output for a given input. The links are associated with a real number that is called the weight of those links. These weights are initialized randomly, and hence there could be a large difference between the actual values and the predicted values. Due to this, it will not give the desired output in one iteration. Even after the weight is assigned and the computation is done if it does not give the desired output, we go back and update the weight of the link with the current value to get closer to the desired output. We do this progressively until we get the best possible output. Also, the weights assigned to the links decides how fast the triggering of the activation function will occur.</p><p>There is another term ‘bias,’ which is used to decide when to trigger the activation function.</p><p>In the real world, the data is always 3-dimensional. For example, let’s say, an image of a car is given as an input to the neural network; we can plot the length and height of the car in a 2-D plane. However, there are a lot more attributes to be considered while computing to recognize it as a car. The computation of this data is a complex task. So, to reduce this difficulty, we use the activation function. It pumps the wide range of values assigned to the data into a specific domain so that computation becomes easy. This is how the artificial neural network works and helps in achieving perfection in Artificial Intelligence.</p><p>To summarize, Artificial Intelligence is an umbrella term, and Machine Learning and Deep Learning are the subdomains of this field that help in achieving Artificial Intelligence. Hopefully, this tutorial gave the hierarchical description of Artificial Intelligence, Machine Learning, and Deep Learning and cleared the confusion among these terms.</p><p><strong>Artificial Neural networks</strong></p><p>ANN was developed considering the same as our brain, the same as how our brain works were taken into account. It was inspired by the way neurons work, the major task is to process information. The architecture of the neural network is similar to neurons. Frank Rosenblatt in 1958 invented ANN and built the machine learning algorithm.</p><p><strong>Neural Network Components</strong></p><p>Each neural network consists of –</p><ul><li>Perceptron — a mathematical representation of the neuron</li><li>Weights, Bias — signifies the importance of each parameter.</li><li>Activation function — at each neuron what makes it to be active or inactive</li><li>Outcome</li><li>Backpropagation</li></ul><p>Data can be of any format — Linear and Nonlinear. The neural networks learn the data types based on the activation function. It can understand the data based on quadratic functions.</p><p><strong>Architecture of Neural network</strong></p><p>It consists of the input value and output value. Each input value is associated with its weight, which passes on to the next level, each perceptron will have an activation function. The weights and input value forms a single perception. We use activation function and based on that, the value goes to next well. And the process continues till it reaches output y’.<br> Sometimes in nonlinear data, the classification is done in three-dimensionality than in two dimensions.<br> <strong>Forward propagation :</strong> <br> The factor 1 keeps moving forward and gets activated in the next level, at the nodes the activation value like sigma, return h, based on that, values like 0, 1 or 2 is passed on to the next level.<br> <strong>Activation functions :</strong><br> These are used to activate neurons. Different activation functions are –<br> The activation function produces the output value 0 or 1, i.e. it classifies the output image for the result.</p><p><strong>Cost function :</strong><br> Sometimes the algorithm we create might predict the value incorrectly, so we need cost function. It tried to quantify the error factor of the neural network. It calculates how well the neural network is performing based on the actual vs predicted value.<br> Error factor = Predicted — Actual.<br> Cost function is used to minimize the error factor.</p><p><strong>Back Propagation :</strong></p><p>When we feel that outputs are not correct, we backpropagate the values to adjust the weights to produce the right output. The architecture, activation functions remains the same in each perceptron. Adjusts using gradient descent. If you have any doubts or queries related to Back Propagation, do post on Artificial Intelligence and Deep Learning community.</p><p><strong>Hyperparameters of ANN :</strong></p><p>1. Hyperparameters are those which are used to tune a neural network. These include –</p><p>2.Learning rate — how fast it abandons the old belief for new ones</p><p>3.Momentum — Smooth learning is maintained by gradient descent.</p><p>4.Epoch — complete forward and backward propagation. As epoch increases, the error decreases.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3594a09abd6" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A beginners to Cloud Computing]]></title>
            <link>https://medium.com/@priyankaattari.pa/a-beginners-to-cloud-computing-97eb94ee3f26?source=rss-9d37d545daa8------2</link>
            <guid isPermaLink="false">https://medium.com/p/97eb94ee3f26</guid>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[interview-questions]]></category>
            <dc:creator><![CDATA[Artificial Intelligence | Salesforce | RPA]]></dc:creator>
            <pubDate>Fri, 08 Nov 2019 12:56:10 GMT</pubDate>
            <atom:updated>2019-11-08T12:56:10.786Z</atom:updated>
            <content:encoded><![CDATA[<p>· With increasing computer and mobile users, data storage has become essential in all fields. Businesses, large or small scale, thrive on their data and spend huge amounts of money to maintain the same. We all have used/experienced cloud computing services in our daily lives, for example, Gmail, Yahoo, Hotmail, etc.</p><p>· The name cloud computing was inspired by the cloud symbol that’s often used to represent the Internet in flowcharts and diagrams as the services provided are available only over the internet.</p><p>· Cloud computing is a system that delivers computing power in the form of CPU, RAM, Storage of software or a service over a network (usually the internet) rather than physically having the computing resources at the customer location. In simple words, It is the practice of using a network of remote servers hosted on the internet to store, manage and process data, rather than a local server or a personal computer. Cloud computing is the delivery of computing services that include servers, storage, databases, networking, software analytics.</p><p>· Cloud computing decreases the hardware and software requirements from the user’s end. The cloud computing systems interface software is accessible through the web browser where the cloud network takes care of the rest.</p><p>· With increasing computer and mobile users, data storage has become essential in all fields. Businesses, large or small scale, thrive on their data and spend huge amounts of money to maintain the same. We all have used/experienced cloud computing services in our daily lives, for example, Gmail, Yahoo, Hotmail, etc.</p><p>· The name cloud computing was inspired by the cloud symbol that’s often used to represent the Internet in flowcharts and diagrams as the services provided are available only over the internet.</p><p>· Cloud computing is a system that delivers computing power in the form of CPU, RAM, Storage of software or a service over a network (usually the internet) rather than physically having the computing resources at the customer location. In simple words, It is the practice of using a network of remote servers hosted on the internet to store, manage and process data, rather than a local server or a personal computer. Cloud computing is the delivery of computing services that include servers, storage, databases, networking, software analytics.</p><p>· Cloud computing decreases the hardware and software requirements from the user’s end. The cloud computing systems interface software is accessible through the web browser where the cloud network takes care of the rest.</p><p><strong>Let’s have a look at the architecture of cloud computing.</strong></p><p>· It consists of two components:</p><p>1. Front end</p><p>consists of the client part of cloud computing. It comprises of interfaces and applications that are required to access the cloud computing platform.</p><p>2. Back end</p><p>Refers to the cloud itself, it comprises of the resources that are essential for the provision of cloud computing services, virtual machines, servers, data storage, security mechanism, etc.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/503/1*FTgfjW65uyGNyag8qc8RwQ.png" /></figure><p>· Cloud computing distributes the file system that spreads over multiple hard disks and machines. Data is never stored in a single location and in case a cloud computing unit fails the other will take over automatically. <strong>Want to become pro in </strong><a href="https://intellipaat.com/all-courses/cloud-computing/"><strong>Cloud Computing Course</strong></a><strong> Click here.</strong></p><p><strong>Types of clouds:</strong></p><p>There are four different cloud models that one can subscribe to according to their needs:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*Lf8P0T-ztt6UOU0uVW_j1g.png" /></figure><p><a href="https://youtu.be/JwmQMi6e9bk">https://youtu.be/JwmQMi6e9bk</a></p><ol><li>Private Cloud:</li></ol><p>Here, computing resources are deployed for one particular organization. This method is more used for intra-business interactions. Where the computing resources can be governed, owned and operated by the same organization.</p><ol><li>Community Cloud:</li></ol><p>Here, computing resources are provided for community and organizations.</p><ol><li>Public Cloud:</li></ol><p>This type of cloud is used usually for B2C (Business to Consumer) type interactions. Here the computing resource is owned, governed and operated by government, an academic or business organization.</p><ol><li>Hybrid Cloud:</li></ol><p>This type of cloud can be used for both types of interactions — B2B (Business to Business) or B2C (Business to Consumer). This deployment method is called a hybrid cloud as the computing resources are bound together by different clouds.</p><p><strong>Cloud Computing Services:</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/848/1*vTT_g15L3hUpjTJxGiwvqg.png" /><figcaption><a href="https://intellipaat.com/all-courses/cloud-computing/">https://intellipaat.com/all-courses/cloud-computing/</a></figcaption></figure><p>There are three major cloud computing services offered</p><ul><li>Software as a Service (SaaS)</li><li>Platform as a Service (PaaS)</li><li>Infrastructure as a Service (IaaS)</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/289/1*Y6Ml3BU2PupoqkrGQd0K7w.png" /></figure><p>Different business uses some or all of these components according to their requirements.</p><p>1. SaaS (Software as a Service):</p><p>§ SaaS or software as a service is a software distribution model in which applications are hosted by a vendor or service provider and made available to customers over a network (internet). SaaS is becoming an increasingly prevalent delivery model as underlying technologies that supports Service Oriented Architecture (SOA) or Web Services. Through the internet, this service is available to users anywhere in the world.</p><p>§ Traditionally, software application needed to be purchased upfront &amp;then installed it onto your computer. SaaS users on the other hand, instead of purchasing the software subscribes to it, usually on a monthly basis via the internet.</p><p>§ Anyone who needs access to a particular piece of software can be subscribed as a user, whether it is one or two people or even thousands of employees in a corporation. SaaS is compatible with all internet-enabled devices. Many important tasks like accounting, sales, invoicing and planning all can be performed using SaaS.</p><ol><li><strong>PaaS (Platform as a Service)</strong></li></ol><p>1. Platform as a service, is referred as PaaS, it provides a platform and environment to allow developers to build applications and services. This service is hosted in the cloud and accessed by the users via the internet.</p><p>2. To understand in simple terms, let compare this with painting a picture, where you are provided with paint colors, different paintbrushes, and paper by your school teacher and you just have to draw a beautiful picture using those tools.</p><p>3. PaaS services are constantly updated &amp; new features added. Software developers, web developers, and businesses can benefit from PaaS. It provides a platform to support application development. It includes software support and management services, storage, networking, deploying, testing, collaborating, hosting and maintaining applications.</p><p><strong>1. IaaS (Infrastructure as a Service)</strong></p><p>a. IaaS (Infrastructure as A Service) is one of the fundamental service model of cloud computing alongside PaaS (Platform as a Service). It provides access to computing resources in a virtualized environment “the cloud” on the internet.</p><p>b. It provides computing infrastructure like virtual server space, network connections, bandwidth, load balancers, and IP addresses. The pool of hardware resources is extracted from multiple servers and networks usually distributed across numerous data centers. This provides redundancy and reliability to IaaS.</p><p>c. IaaS (Infrastructure as a service) is a complete package for computing. For small scale businesses that are looking for cutting costs on IT infrastructure, IaaS is one of the solutions. Annually a lot of money is spent on maintenance and buying new components like hard-drives, network connections, external storage devices, etc. which a business owner could have saved for other expenses by using IaaS.</p><p><strong>Security Concerns for cloud computing:</strong></p><p>While using cloud computing, the major issue that concerns users is about its security and privacy. To provide security for systems, networks and data cloud computing service providers have joined hands with TCG (Trusted Computing Group) which is a non-profit organization that regularly releases a set of specifications to secure hardware, create self-encrypting drives and improve network security. It protects the data from rootkits and malware.</p><p>Privacy is a strong barrier for users to inculcate into cloud computing systems. To improve privacy in cloud computing certain measures should be adopted.</p><p>· Theoretically monitoring data movement in memory before it is stored in a disk. To keep the confidentiality of the data, administrative and legal controls should prevent this from happening.</p><p>· Keeping the data encrypted at the cloud storage site, preventing unauthorized access through the internet, preventing cloud vendors to access the data too.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=97eb94ee3f26" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>