<?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 Robin Reni on Medium]]></title>
        <description><![CDATA[Stories by Robin Reni on Medium]]></description>
        <link>https://medium.com/@robinandreaureni?source=rss-d54d7dc78c10------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*_0sHF71APtsZlgdxXkZLxg.jpeg</url>
            <title>Stories by Robin Reni on Medium</title>
            <link>https://medium.com/@robinandreaureni?source=rss-d54d7dc78c10------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 22:26:10 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@robinandreaureni/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[GENESIS OF AI]]></title>
            <link>https://medium.com/@robinandreaureni/genesis-of-ai-b6e1c77dfcaf?source=rss-d54d7dc78c10------2</link>
            <guid isPermaLink="false">https://medium.com/p/b6e1c77dfcaf</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[cognitive-computing]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[deep-learning]]></category>
            <dc:creator><![CDATA[Robin Reni]]></dc:creator>
            <pubDate>Tue, 08 May 2018 20:20:17 GMT</pubDate>
            <atom:updated>2018-05-08T20:20:17.811Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/712/1*wykqQy2AID2FPpqYm0A7lA.jpeg" /></figure><p>Now we are in the Fourth Industrial Revolution called THE AI REVOLUTION. Things are getting changed up and more precisely they are getting automated.AI had stamped its feet on all sectors of industries most importantly on Healthcare, Finance, Transportation etc. More research is going on to enhance AI ability and their functionalities. Remember AI is not a single field of a subject it includes a variety of core subjects which are integrated to bring automation into this world.</p><p>Don’t worry this blog is full of General Knowledge and cool applications of AI not any technical stuff into it.</p><p>Many of advancement of AI is going on a Business angle like How to make money out of it? and How can we enhance our product by using AI in it ?. Only giant companies like Google, OpenAI, Microsoft etc are really working on non-profit AI products. In this blog, we gonna see the top application of AI other than Business domain. Really the topics gonna be an awesome one.</p><h4>1.MUSIC :</h4><p><strong>” Music is one of the greater uniter and an incredible force something that people who differ on everything and anything else have common. — Sarah Dessen ”</strong></p><p>Music is one of the Stressbuster of every Human Beings.No person in the world is there without listening music. Music is created by a sequence of sound from various instruments and voice. All the sound should be set on a rhythm to maintain the flow of music. But till 2016 only human had the ability to set the rhythm to give a quality music to the world. Then AI grabbed the chance of it, Google created a project called <strong>Magenta</strong> which gives all the technical support to create AI-powered and <strong>AIVA </strong>is also music product based company giving quality AI music to this world.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/480/0*8MZqrPuqpSsVDX0u." /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/468/0*ZQJi-NDHKD_ilf1l." /></figure><h4>2.SPACE :</h4><p><strong>” I don’t think the human race will survive the next thousand years unless we spread into space — Stephen Hawkings ”</strong></p><p>Space exploration had become a much-needed necessity for the human kind. Earth is never a permanent place human have to search another solution instead of Earth. Many Space missions like <strong>Mars Mission , Reusable Rockets , Colonization of Mars, Black hole discovery</strong> etc have been launched by<strong> NASA , SpaceX, ISRO</strong> etc to advance the space exploration level.You will have the question really<strong> AI have the power to expolore the universe ?</strong> The anwer is partially AI has the power to virtualize the universe and explore it . NASA is using some advanced AI for communications , self-exploration , autonoumous vehicle etc .NASA<strong> FDL</strong> is one of the famous project powered by AI for space exploration. <strong>ETH ZURICH</strong> <strong>COLLEGE</strong> has also contriubuted for space exploration using AI.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/545/0*9hAfCs-la7tAN0lZ." /></figure><h4>3.ART :</h4><p><strong>” Every child is an artist ,the probelm is staying an artist when you grow up — Pablo Picasso “</strong></p><p>Art is creative activity which describes ones imagination in a real form .All of them are basically a artist , but the thing is how excellent we are to express it . A single piece of art have many projection and many meanings into it .Only a perfect artist can really do that . <strong>How about AI are they really a good artist ? </strong>Yes , they do really do amazing master piece like human.<strong>Pindar Van Arman ,CEO of Cloudpainter</strong> do amazing artist work using AI .<strong>The Art and Artifical Intelligence Lab</strong> also conducting various research on arts using AI.<strong>The Painting Fool</strong> is also a commercial company selling AI developed arts.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/544/0*4IvjSlOB8-z_gyN8." /></figure><h4>4.BOARD GAMES :</h4><p><strong>“Change the game , dont let the game Change You — Macklemore”</strong></p><p>Now I am not gonna talk about computer games or software oriented.I am gonna explain about the real masterstroke games like <strong>Chess, Alpha go, Othello</strong> etc. Games are the system which checks the competency level of each other. Huge level of dedication and hard work are required to be a champion in a game. <strong>AlphaGo</strong>, one of the worlds ancient and challenging game. Lee Sidol, 18 international titles on AlphaGo championship.<strong> Do you believe that an AI beat him?</strong> Yes, One of the Google famous projects was <strong>Deepminds AlphaGo</strong>, they developed an AI which has the ability to beat the world champion. Considerably<strong> IBM Deepblue</strong> also titles the first AI Chess Champion. Scientist says that Gaming algorithm for AI is the best solution to solve a critical problem and also increase its computational capability.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/501/0*cRe4ruEwRs6Qw-Ge." /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/508/0*YRvr73uFGsaQvHkn." /></figure><h4>5.CONTENT CREATOR :</h4><p><strong>“Writing is the Painting of our voice “</strong></p><p>Majority of human beings love to read books, essays, content etc. But only a few like to write those contents. As a content writer, they should know all the subject depth and its theories to give a perfect one. Many of the contents are liked and disliked. So the authors should have some understanding of the human expectation and create according to that. <strong>Does AI understands Human Expectation and create content Accordingly?</strong> Yes, it can. <strong>Japanese AI won the Best Novel prize for the novel “The Day a Computer Writes a Novel”</strong>. <strong>AI Writer</strong> an emerging company writing content using AI for free. There is some assumption that AI can write novels and content better than human in future . Let us see for the perfect novel by an AI.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/990/0*x-1dNy0SucPDw4s7.jpg" /></figure><p>That’s it. These are some of the cool stuff of AI out of the Business box and there are much more like this. AI is generally attracting humans indirectly by its functionalities and one-day human race will face the entire work of it. The day can be worse one or a life-changing one.</p><p>So you decide AI is a gift or a curse to the mankind.</p><p>I hope you like this blog. Feel free to post your comments and support the blog by your like. Happy Coding 🙂</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/850/1*qnQUY2_gyWxo8xsTT0kzQA.jpeg" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b6e1c77dfcaf" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[MACHINE LEARNING IN ACTION]]></title>
            <link>https://medium.com/@robinandreaureni/machine-learning-in-action-52886ef17dfb?source=rss-d54d7dc78c10------2</link>
            <guid isPermaLink="false">https://medium.com/p/52886ef17dfb</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Robin Reni]]></dc:creator>
            <pubDate>Mon, 08 Jan 2018 21:09:03 GMT</pubDate>
            <atom:updated>2018-01-08T21:09:03.072Z</atom:updated>
            <content:encoded><![CDATA[<p>Machine Learning (ML) is coming into its own, with a growing recognition that ML can play a key role in a wide range of critical applications, such as data mining, natural language processing, image recognition, and expert systems. ML provides potential solutions in all these domains and more, and is set to be a pillar of our future civilization.</p><p>The supply of <a href="https://www.toptal.com/machine-learning">able ML designers</a> has yet to catch up to this demand. A major reason for this is that ML is just plain tricky. This tutorial introduces the basics of Machine Learning theory, laying down the common themes and concepts, making it easy to follow the logic and get comfortable with the topic.</p><h3>What is Machine Learning?</h3><p>So what exactly is “machine learning” anyway? ML is actually a <em>lot</em> of things. The field is quite vast and is expanding rapidly, being continually partitioned and sub-partitioned ad nauseam into different sub-specialties and <a href="https://www.toptal.com/machine-learning/an-introduction-to-deep-learning-from-perceptrons-to-deep-networks">types of machine learning</a>.</p><p>There are some basic common threads, however, and the overarching theme is best summed up by this oft-quoted statement made by Arthur Samuel way back in 1959: <em>“[Machine Learning is the] field of study that gives computers the ability to learn without being explicitly programmed.”</em></p><p>And more recently, in 1997, <a href="http://www.cs.cmu.edu/~tom/">Tom Mitchell</a> gave a “well-posed” definition that has proven more useful to engineering types: <em>“A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E.”</em></p><p>“A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E.” — Tom Mitchell, Carnegie Mellon University</p><p>So if you want your program to predict, for example, traffic patterns at a busy intersection (task T), you can run it through a machine learning algorithm with data about past traffic patterns (experience E) and, if it has successfully “learned”, it will then do better at predicting future traffic patterns (performance measure P).</p><p>The highly complex nature of many real-world problems, though, often means that inventing specialized algorithms that will solve them perfectly every time is impractical, if not impossible. Examples of machine learning problems include, <a href="http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2675494/">“Is this cancer?”</a>, <a href="http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1316046">“What is the market value of this house?”</a>, <a href="http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2187186">“Which of these people are good friends with each other?”</a>, <a href="https://archive.org/stream/nasa_techdoc_19960011791/19960011791_djvu.txt">“Will this rocket engine explode on take off?”</a>, <a href="http://techblog.netflix.com/2012/04/netflix-recommendations-beyond-5-stars.html">“Will this person like this movie?”</a>, <a href="http://www.fastcolabs.com/3028414/how-facebooks-machines-got-so-good-at-recognizing-your-face">“Who is this?”</a>, <a href="https://en.wikipedia.org/wiki/Natural_language_processing#NLP_using_machine_learning">“What did you say?”</a>, and <a href="http://phys.org/news139501522.html">“How do you fly this thing?”</a>. All of these problems are excellent targets for an ML project, and in fact ML has been applied to each of them with great success.</p><p>ML solves problems that cannot be solved by numerical means alone.</p><p>Among the different types of ML tasks, a crucial distinction is drawn between supervised and unsupervised learning:</p><ul><li><strong>Supervised machine learning:</strong> The program is “trained” on a pre-defined set of “training examples”, which then facilitate its ability to reach an accurate conclusion when given new data.</li><li><strong>Unsupervised machine learning:</strong> The program is given a bunch of data and must find patterns and relationships therein.</li></ul><p>We will primarily focus on supervised learning here, but the end of the article includes a brief discussion of unsupervised learning with some links for those who are interested in pursuing the topic further.</p><h3>Supervised Machine Learning</h3><p>In the majority of supervised learning applications, the ultimate goal is to develop a finely tuned predictor function h(x) (sometimes called the “hypothesis”). “Learning” consists of using sophisticated mathematical algorithms to optimize this function so that, given input data x about a certain domain (say, square footage of a house), it will accurately predict some interesting value h(x) (say, market price for said house).</p><p>In practice, x almost always represents multiple data points. So, for example, a housing price predictor might take not only square-footage (x1) but also number of bedrooms (x2), number of bathrooms (x3), number of floors (x4), year built (x5), zip code (x6), and so forth. Determining which inputs to use is an important part of ML design. However, for the sake of explanation, it is easiest to assume a single input value is used.</p><p>So let’s say our simple predictor has this form:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/169/0*_nt83EywQTjPfgjK.png" /></figure><p>where</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*6AV2qihwMIMztqDq.png" /></figure><p>and</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*ZeZbE6UFPru5yYg6.png" /></figure><p>are constants. Our goal is to find the perfect values of</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*8f2CroH2X-QCdsy8.png" /></figure><p>and</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*UV1Olc-Eeykc4MhH.png" /></figure><p>to make our predictor work as well as possible.</p><p>Optimizing the predictor h(x) is done using <strong>training examples</strong>. For each training example, we have an input value x_train, for which a corresponding output, y, is known in advance. For each example, we find the difference between the known, correct value y, and our predicted value h(x_train). With enough training examples, these differences give us a useful way to measure the “wrongness” of h(x). We can then tweak h(x) by tweaking the values of</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*gJkBj1e4GBsQbylx.png" /></figure><p>and</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*z8eloYadNY8gXwJo.png" /></figure><p>to make it “less wrong”. This process is repeated over and over until the system has converged on the best values for</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*rMLNHyUUfOb2V81S.png" /></figure><p>and</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*grtLSL9G8sDdy1pE.png" /></figure><p>. In this way, the predictor becomes trained, and is ready to do some real-world predicting.</p><h3>A Simple Machine Learning Example</h3><p>We stick to simple problems in this post for the sake of illustration, but the reason ML exists is because, in the real world, the problems are much more complex. On this flat screen we can draw you a picture of, at most, a three-dimensional data set, but ML problems commonly deal with data with millions of dimensions, and very complex predictor functions. ML solves problems that cannot be solved by numerical means alone.</p><p>With that in mind, let’s look at a simple example. Say we have the following training data, wherein company employees have rated their satisfaction on a scale of 1 to 100:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*OFLxh492XH0uCtu0.png" /></figure><p>First, notice that the data is a little noisy. That is, while we can see that there is a pattern to it (i.e. employee satisfaction tends to go up as salary goes up), it does not all fit neatly on a straight line. This will always be the case with real-world data (and we absolutely want to train our machine using real-world data!). So then how can we train a machine to perfectly predict an employee’s level of satisfaction? The answer, of course, is that we can’t. The goal of ML is never to make “perfect” guesses, because ML deals in domains where there is no such thing. The goal is to make guesses that are good enough to be useful.</p><p>It is somewhat reminiscent of the famous statement by British mathematician and professor of statistics <a href="https://en.wikipedia.org/wiki/George_E._P._Box">George E. P. Box</a> that “all models are wrong, but some are useful”.</p><p>The goal of ML is never to make “perfect” guesses, because ML deals in domains where there is no such thing. The goal is to make guesses that are good enough to be useful.</p><p>ML builds heavily on statistics. For example, when we train our machine to learn, we have to give it a statistically significant random sample as training data. If the training set is not random, we run the risk of the machine learning patterns that aren’t actually there. And if the training set is too small (see <a href="https://en.wikipedia.org/wiki/Law_of_large_numbers">law of large numbers</a>), we won’t learn enough and may even reach inaccurate conclusions. For example, attempting to predict company-wide satisfaction patterns based on data from upper management alone would likely be error-prone.</p><p>With this understanding, let’s give our machine the data we’ve been given above and have it learn it. First we have to initialize our predictor h(x) with some reasonable values of</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*Qc0ZOlx-JOqlNR7x.png" /></figure><p>and</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*NqrEfiLUwx6Jx-Kl.png" /></figure><p>. Now our predictor looks like this when placed over our training set:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/224/0*rIp2DWsrJvP3610x.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*3aWC4661CUO17mzG.png" /></figure><p>If we ask this predictor for the satisfaction of an employee making $60k, it would predict a rating of 27:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*i9v2dBHpzHxeO3KN.png" /></figure><p>It’s obvious that this was a terrible guess and that this machine doesn’t know very much.</p><p>So now, let’s give this predictor <em>all</em> the salaries from our training set, and take the differences between the resulting predicted satisfaction ratings and the actual satisfaction ratings of the corresponding employees. If we perform a little mathematical wizardry (which I will describe shortly), we can calculate, with very high certainty, that values of 13.12 for</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*SJi_UI4iAlsBDWMx.png" /></figure><p>and 0.61 for</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*fECUJjSHLpJziHrM.png" /></figure><p>are going to give us a better predictor.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/224/0*O7sjk0VTqoinB5TI.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*EhPEwJ3ItKhccIxw.png" /></figure><p>And if we repeat this process, say 1500 times, our predictor will end up looking like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/224/0*pR9fYNZgam9ExsMJ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*onagq0naiCRIIZ1d.png" /></figure><p>At this point, if we repeat the process, we will find that</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*g3Iv9XnwVBgh_QYO.png" /></figure><p>and</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*R6oU_n94r7m3NVRR.png" /></figure><p>won’t change by any appreciable amount anymore and thus we see that the system has converged. If we haven’t made any mistakes, this means we’ve found the optimal predictor. Accordingly, if we now ask the machine again for the satisfaction rating of the employee who makes $60k, it will predict a rating of roughly 60.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*62WurJBZcZ1pIEAS.png" /></figure><p>Now we’re getting somewhere.</p><h3>A Note on Complexity</h3><p>The above example is technically a simple problem of <a href="http://www.mathworks.com/help/symbolic/mupad_ug/univariate-linear-regression.html">univariate linear regression</a>, which in reality can be solved by deriving a simple normal equation and skipping this “tuning” process altogether. However, consider a predictor that looks like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/584/0*czp7FfUokRExnC1y.png" /></figure><p>This function takes input in four dimensions and has a variety of polynomial terms. Deriving a normal equation for this function is a significant challenge. Many modern machine learning problems take thousands or even millions of dimensions of data to build predictions using hundreds of coefficients. Predicting how an organism’s genome will be expressed, or what the climate will be like in fifty years, are examples of such complex problems.</p><p>Many modern ML problems take thousands or even millions of dimensions of data to build predictions using hundreds of coefficients.</p><p>Fortunately, the iterative approach taken by ML systems is much more resilient in the face of such complexity. Instead of using brute force, a machine learning system “feels its way” to the answer. For big problems, this works much better. While this doesn’t mean that ML can solve all arbitrarily complex problems (it can’t), it does make for an incredibly flexible and powerful tool.</p><h3>Gradient Descent — Minimizing “Wrongness”</h3><p>Let’s take a closer look at how this iterative process works. In the above example, how do we make sure</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*MtVT--VqjCU4foqe.png" /></figure><p>and</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*6hlfWt6LosGMuT6U.png" /></figure><p>are getting better with each step, and not worse? The answer lies in our “measurement of wrongness” alluded to previously, along with a little calculus.</p><p>The wrongness measure is known as the <strong>cost function</strong> (a.k.a., <strong>loss function</strong>),</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/30/0*neDyNBNhe9gPOhEc.png" /></figure><p>. The input</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/10/0*Epl04ALbf8g49glZ.png" /></figure><p>represents all of the coefficients we are using in our predictor. So in our case,</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/10/0*b4tYxA07o2MTFFB8.png" /></figure><p>is really the pair</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*qAGHCZQ3lKFoS72o.png" /></figure><p>and</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*gKDY8cgst9E8PGte.png" /></figure><p>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/60/0*GEjwBM-gzhmaIttH.png" /></figure><p>gives us a mathematical measurement of how wrong our predictor is when it uses the given values of</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*ogu3V2aGzepEuyNo.png" /></figure><p>and</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*TLIfPdQ0t9X5mVsv.png" /></figure><p>.</p><p>The choice of the cost function is another important piece of an ML program. In different contexts, being “wrong” can mean very different things. In our employee satisfaction example, the well-established standard is the <a href="https://en.wikipedia.org/wiki/Linear_least_squares_(mathematics)">linear least squares function</a>:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/307/0*I4YcpZdjR_mlZ0S4.png" /></figure><p>With least squares, the penalty for a bad guess goes up quadratically with the difference between the guess and the correct answer, so it acts as a very “strict” measurement of wrongness. The cost function computes an average penalty over all of the training examples.</p><p>So now we see that our goal is to find</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*PkG7zTWQnZUHcYYT.png" /></figure><p>and</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*0gNSFlNHF3g00csE.png" /></figure><p>for our predictor h(x) such that our cost function</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/60/0*5LtF75yizK0kJkD2.png" /></figure><p>is as small as possible. We call on the power of calculus to accomplish this.</p><p>Consider the following plot of a cost function for some particular ML problem:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*cxPvlieE4uXK2fVt.png" /></figure><p>Here we can see the cost associated with different values of</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*w4Tkl15N0RhuSsgS.png" /></figure><p>and</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*0FoK5qdj1G6dWXKX.png" /></figure><p>. We can see the graph has a slight bowl to its shape. The bottom of the bowl represents the lowest cost our predictor can give us based on the given training data. The goal is to “roll down the hill”, and find</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*bC3WGAcTNlXREyd6.png" /></figure><p>and</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*orvpOsqPCJ-ZvLaf.png" /></figure><p>corresponding to this point.</p><p>This is where calculus comes in to this machine learning tutorial. For the sake of keeping this explanation manageable, I won’t write out the equations here, but essentially what we do is take the gradient of</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/60/0*zz0ZFiRnaLKuxBUk.png" /></figure><p>, which is the pair of derivatives of</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/60/0*LqlFBkroBfOBKbxk.png" /></figure><p>(one over</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*YLfnl0bDKgmAVygO.png" /></figure><p>and one over</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*tGKr1yl0vr17LXEW.png" /></figure><p>). The gradient will be different for every different value of</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*P8XSAXetagRkP4WV.png" /></figure><p>and</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*QcKcJ8j6U6Ey7pb_.png" /></figure><p>, and tells us what the “slope of the hill is” and, in particular, “which way is down”, for these particular</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/10/0*wxLNnO54X-FbJ6kA.png" /></figure><p>s. For example, when we plug our current values of</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/10/0*dC19ShGbWkwqA3ex.png" /></figure><p>into the gradient, it may tell us that adding a little to</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*2O6rqKEC-0e76JD0.png" /></figure><p>and subtracting a little from</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*qrkXZKlybAnC0V13.png" /></figure><p>will take us in the direction of the cost function-valley floor. Therefore, we add a little to</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*SD9NUioDBFyXfLqX.png" /></figure><p>, and subtract a little from</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*FZWDR8kcqFjg5L1n.png" /></figure><p>, and voilà! We have completed one round of our learning algorithm. Our updated predictor, h(x) =</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/15/0*XkbxAC4iDdvknnet.png" /></figure><p>+</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/14/0*Y8P1zz3gCQU6fbFB.png" /></figure><p>x, will return better predictions than before. Our machine is now a little bit smarter.</p><p>This process of alternating between calculating the current gradient, and updating the</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/10/0*DolkleNvYVVe10fq.png" /></figure><p>s from the results, is known as <a href="https://en.wikipedia.org/wiki/Gradient_descent">gradient descent</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*ClkaE9Hjg-Hjmuhh.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*RSLUj2--44LGZ3ba.png" /></figure><p>That covers the basic theory underlying the majority of supervised Machine Learning systems. But the basic concepts can be applied in a variety of different ways, depending on the problem at hand.</p><h3>Classification Problems</h3><p>Under supervised ML, two major subcategories are:</p><ul><li><strong>Regression machine learning systems:</strong> Systems where the value being predicted falls somewhere on a continuous spectrum. These systems help us with questions of “How much?” or “How many?”.</li><li><strong>Classification machine learning systems:</strong> Systems where we seek a yes-or-no prediction, such as “Is this tumer cancerous?”, “Does this cookie meet our quality standards?”, and so on.</li></ul><p>As it turns out, the underlying theory is more or less the same. The major differences are the design of the predictor h(x) and the design of the cost function</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/30/0*U0JsTI3KWgNH6bEe.png" /></figure><p>.</p><p>Our examples so far have focused on regression problems, so let’s now also take a look at a classification example.</p><p>Here are the results of a cookie quality testing study, where the training examples have all been labeled as either “good cookie” (y = 1) in blue or “bad cookie” (y = 0) in red.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*hzQAnNT_t_jcrSuA.png" /></figure><p>In classification, a regression predictor is not very useful. What we usually want is a predictor that makes a guess somewhere between 0 and 1. In a cookie quality classifier, a prediction of 1 would represent a very confident guess that the cookie is perfect and utterly mouthwatering. A prediction of 0 represents high confidence that the cookie is an embarrassment to the cookie industry. Values falling within this range represent less confidence, so we might design our system such that prediction of 0.6 means “Man, that’s a tough call, but I’m gonna go with yes, you can sell that cookie,” while a value exactly in the middle, at 0.5, might represent complete uncertainty. This isn’t always how confidence is distributed in a classifier but it’s a very common design and works for purposes of our illustration.</p><p>It turns out there’s a nice function that captures this behavior well. It’s called the <a href="https://en.wikipedia.org/wiki/Sigmoid_function">sigmoid function</a>, g(z), and it looks something like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/124/0*BdhdcMFhUOhllfDT.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*9XoMIjBcph_Lz9r8.png" /></figure><p>z is some representation of our inputs and coefficients, such as:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/134/0*1loTfKYdWMh-ZRfC.png" /></figure><p>so that our predictor becomes:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/181/0*3jSIWRNGFgCinlfo.png" /></figure><p>Notice that the sigmoid function transforms our output into the range between 0 and 1.</p><p>The logic behind the design of the cost function is also different in classification. Again we ask “what does it mean for a guess to be wrong?” and this time a very good rule of thumb is that if the correct guess was 0 and we guessed 1, then we were completely and utterly wrong, and vice-versa. Since you can’t be more wrong than absolutely wrong, the penalty in this case is enormous. Alternatively if the correct guess was 0 and we guessed 0, our cost function should not add any cost for each time this happens. If the guess was right, but we weren’t completely confident (e.g. y = 1, but h(x) = 0.8), this should come with a small cost, and if our guess was wrong but we weren’t completely confident (e.g. y = 1 but h(x) = 0.3), this should come with some significant cost, but not as much as if we were completely wrong.</p><p>This behavior is captured by the log function, such that:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/355/0*Kbd8S3WWUi9SjRAr.png" /></figure><p>Again, the cost function</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/30/0*dhNFICOXWVs-s8wc.png" /></figure><p>gives us the average cost over all of our training examples.</p><p>So here we’ve described how the predictor h(x) and the cost function</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/30/0*07knO3ThLyCQ6hgC.png" /></figure><p>differ between regression and classification, but gradient descent still works fine.</p><p>A classification predictor can be visualized by drawing the boundary line; i.e., the barrier where the prediction changes from a “yes” (a prediction greater than 0.5) to a “no” (a prediction less than 0.5). With a well-designed system, our cookie data can generate a classification boundary that looks like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*VfqxwT3Z8MmgxHRP.png" /></figure><p>Now that’s a machine that knows a thing or two about cookies!</p><h3>An Introduction to Neural Networks</h3><p>No discussion of ML would be complete without at least mentioning <a href="https://en.wikipedia.org/wiki/Artificial_neural_network">neural networks</a>. Not only do neural nets offer an extremely powerful tool to solve very tough problems, but they also offer fascinating hints at the workings of our own brains, and intriguing possibilities for one day creating truly intelligent machines.</p><p>Neural networks are well suited to machine learning problems where the number of inputs is gigantic. The computational cost of handling such a problem is just too overwhelming for the types of systems we’ve discussed above. As it turns out, however, neural networks can be effectively tuned using techniques that are strikingly similar to gradient descent in principle.</p><h3>Unsupervised Machine Learning</h3><p>Unsupervised learning typically is tasked with finding relationships within data. There are no training examples used in this process. Instead, the system is given a set data and tasked with finding patterns and correlations therein. A good example is identifying close-knit groups of friends in social network data.</p><p>The algorithms used to do this are very different from those used for supervised learning, and the topic merits its own post. However, for something to chew on in the meantime, take a look at <a href="https://en.wikipedia.org/wiki/Cluster_analysis">clustering algorithms</a> such as <a href="https://en.wikipedia.org/wiki/K-means_clustering">k-means</a>, and also look into <a href="https://en.wikipedia.org/wiki/Dimensionality_reduction">dimensionality reduction</a> systems such as <a href="https://en.wikipedia.org/wiki/Principal_component_analysis">principle component analysis</a>. Our prior <a href="https://www.toptal.com/big-data#hiring-guide">post on big data</a> discusses a number of these topics in more detail as well.</p><h3>Conclusion</h3><p>We’ve covered much of the basic theory underlying the field of Machine Learning here, but of course, we have only barely scratched the surface.</p><p>Keep in mind that to really apply the theories contained in this introduction to real life machine learning examples, a much deeper understanding of the topics discussed herein is necessary. There are many subtleties and pitfalls in ML, and many ways to be lead astray by what appears to be a perfectly well-tuned thinking machine. Almost every part of the basic theory can be played with and altered endlessly, and the results are often fascinating. Many grow into whole new fields of study that are better suited to particular problems.</p><p>Clearly, Machine Learning is an incredibly powerful tool. In the coming years, it promises to help solve some of our most pressing problems, as well as open up whole new worlds of opportunity. The <a href="https://www.toptal.com/machine-learning">demand for ML engineers</a> is only going to continue to grow, offering incredible chances to be a part of something big. I hope you will consider getting in on the action!</p><h3>Acknowledgement</h3><p>This article draws heavily on material taught by Stanford Professor <a href="http://cs.stanford.edu/people/ang/">Dr. Andrew Ng</a> in his <a href="https://www.coursera.org/course/ml">free and open Machine Learning course</a>. The course covers everything discussed in this article in great depth and gives tons of practical advice for the ML practitioner. I cannot recommend this course highly enough for those interested in further exploring this fascinating field.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=52886ef17dfb" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Getting started with AI]]></title>
            <link>https://medium.com/buzzrobot/getting-started-with-ai-7ddffa2a586?source=rss-d54d7dc78c10------2</link>
            <guid isPermaLink="false">https://medium.com/p/7ddffa2a586</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[computer-science]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Robin Reni]]></dc:creator>
            <pubDate>Mon, 08 Jan 2018 20:47:12 GMT</pubDate>
            <atom:updated>2018-01-18T21:35:46.454Z</atom:updated>
            <content:encoded><![CDATA[<p>One of the breakthrough of Technology in 21st Century is Artificial Intelligence.It started to adapt various fields in the industry and became an emerging technology.So that there is a huge demand for AI Developers and Researcher and many of the organization and individuals contributed their work in AI which leads to the topic called Machine learning and Deep Learning.Many of the Developer are choosing AI as their career but they don’t know How to Start Learning AI ?. So in this blog, I am going to share some information and resources to learn AI.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*_IPoH8Az_Dl1DyJ0.png" /></figure><p>The first Step Learn Difference Between AI, Machine Learning, and Deep Learning:</p><p><strong>Artificial Intelligence:</strong></p><p>Any techniques that enable the computer to mimic human intelligence using logic,if-then rules, Decision trees and machine learning including Deep learning is called Artificial Intelligence.</p><p>Eg: Google Assistant</p><p><strong>Machine Learning:</strong></p><p>A subset of AI that includes abstruse statistical techniques that enable machines to improve tasks with experience.The category includes Deep Learning.</p><p>Eg: Predicting the house rate by the past available data</p><p><strong>Deep Learning:</strong></p><p>The subset of Machine Learning composed of Algorithms that permit software to train itself to perform tasks like speech and image recognition by exposing multi-layered neural networks to vast amounts of data.</p><p>Eg: Object Detection whether its a cat or dog using pre-trained models.</p><h3>Programming Knowledge For AI :</h3><ul><li>Python (Since it has many community support and framework to develop efficient AI programs).To get started with python refer the below link</li></ul><p><a href="https://www.programiz.com/python-programming">https://www.programiz.com/python-programming</a></p><p>Python Books:</p><p><a href="https://mega.nz/#F!jghEBTKJ!KK7R3EoVBSngO5lK5RWEFw">https://mega.nz/#F!jghEBTKJ!KK7R3EoVBSngO5lK5RWEFw</a></p><p>Python Documentation:</p><p><a href="https://docs.python.org/3/">https://docs.python.org/3/</a></p><ul><li>Little Bit of SQL:</li></ul><p><a href="https://www.w3schools.com/sql/default.asp">https://www.w3schools.com/sql/default.asp</a></p><p>That’s it, I think this much is enough to get started.</p><h3>REMEMBER:</h3><p><strong>While watching videos kindly make the note of important keywords, if you didn’t understand watch it again or browse it or comment on my blog I will help you.Please don’t SKIP.</strong></p><h3>Concepts Of AI:</h3><p>When you are going to start studying about AI you should be well clear about their terminology and their algorithms.Because different use cases working with different AI algorithm.Eg : For prototyping, any game powered by AI most of them will be using <strong>MIN-MAX ALGORITHM. </strong>So you should be clear with AI concept when you are going to do a project in that.I can give you some major algorithm and concepts and their link below:</p><ul><li>Intelligent Agents and Uninformed Search</li><li>Heuristic Search</li><li>Adversarial Search and Games</li><li>Machine Learning</li><li>Constraint Satisfaction problems</li><li>Reinforcement Learning</li><li>Logical Agents</li><li>Natural Language Processing</li><li>Deep Learning</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/0*WD73knM7zatSvwKR.jpg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*WbJXI0CweYnE2gbl.jpg" /></figure><p>It will be theory part so make your mindset according to that and start learning it.The useful links are:</p><ul><li><a href="https://courses.edx.org/courses/course-v1:ColumbiaX+CSMM.101x+2T2017_2/course/">https://courses.edx.org/courses/course-v1:ColumbiaX+CSMM.101x+2T2017_2/course/</a></li><li><a href="https://in.udacity.com/course/intro-to-artificial-intelligence--cs271">https://in.udacity.com/course/intro-to-artificial-intelligence--cs271</a></li></ul><p>I recommend these two are the best beginner course for AI and its FREE.</p><h3>PICK UP THE CORRECT CATEGORY IN AI:</h3><p>After reading the previous sub-topic you should be clear with the concepts of it.Now it’s the time to pick the right category from it.Since to make my blog short and brief I will discuss only the emerging category of AI by their popularity.</p><h4>1.Machine Learning:</h4><p>Machine learning is the science of getting computers to act without being explicitly programmed. In the past decade, machine learning has given us self-driving cars, practical speech recognition, effective web search, and a vastly improved understanding of the human genome. Machine learning is so pervasive today that you probably use it dozens of times a day without knowing it. Many researchers also think it is the best way to make progress towards human-level AI.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/422/0*x1US0TgejfJOskvn.PNG" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*0ZbAfGapht7UrDIU.png" /></figure><p>Useful Course Links and Blogs for Machine Learning:</p><ul><li><a href="https://www.kdnuggets.com/2016/08/10-algorithms-machine-learning-engineers.html">https://www.kdnuggets.com/2016/08/10-algorithms-machine-learning-engineers.html</a></li><li><a href="https://classroom.udacity.com/courses/ud120">https://classroom.udacity.com/courses/ud120</a></li><li><a href="https://www.coursera.org/learn/machine-learning">https://www.coursera.org/learn/machine-learning</a></li><li><a href="https://medium.com/machine-learning-for-humans/why-machine-learning-matters-6164faf1df12">https://medium.com/machine-learning-for-humans/why-machine-learning-matters-6164faf1df12</a></li><li><a href="https://www.youtube.com/watch?v=HcqpanDadyQ">https://www.youtube.com/watch?v=HcqpanDadyQ</a></li><li><a href="https://youtu.be/OGxgnH8y2NM">https://youtu.be/OGxgnH8y2NM</a></li></ul><p>I think this much is enough to get started with ML and mostly you will be covering all the major topics of it.</p><p>BOOKS FOR REFERENCE:</p><ul><li><a href="http://shop.oreilly.com/product/0636920030515.do">http://shop.oreilly.com/product/0636920030515.do</a></li><li><a href="http://shop.oreilly.com/product/0636920052289.do">http://shop.oreilly.com/product/0636920052289.do</a></li><li><a href="https://www.manning.com/books/machine-learning-in-action">https://www.manning.com/books/machine-learning-in-action</a></li></ul><p>BLOG LINKS FOR ML:</p><ul><li><a href="https://aitoml.wordpress.com/2017/10/29/tips-about-machine-learning/">https://aitoml.wordpress.com/2017/10/29/tips-about-machine-learning/</a></li><li><a href="https://aitoml.wordpress.com/2017/10/29/machine-learning-introductory-tutorial/">https://aitoml.wordpress.com/2017/10/29/machine-learning-introductory-tutorial/</a></li><li><a href="https://aitoml.wordpress.com/2017/10/29/mathematics-topics-for-machine-learning/">https://aitoml.wordpress.com/2017/10/29/mathematics-topics-for-machine-learning/</a></li></ul><p><strong>2.Natural Language Processing:</strong></p><p><strong>Natural language processing</strong> (<strong>NLP</strong>) is a field of computer science, artificial intelligence concerned with the interactions between computers and human (natural) languages, and, in particular, concerned with programming computers to fruitfully process large natural language data.Challenges in natural language processing frequently involve speech recognition, natural language understanding, and natural language generation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/638/0*fRSrn4fOPe_Me4gP." /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*GA_bV4w4cImdT7aU.png" /></figure><p>Links For NLP:</p><ul><li><a href="https://www.edx.org/course/natural-language-processing-nlp-microsoft-dev288x">https://www.edx.org/course/natural-language-processing-nlp-microsoft-dev288x</a></li><li><a href="https://www.coursera.org/learn/natural-language-processing">https://www.coursera.org/learn/natural-language-processing</a></li><li><a href="http://web.stanford.edu/class/cs224n/">http://web.stanford.edu/class/cs224n/</a></li><li><a href="https://youtu.be/FLZvOKSCkxY">https://youtu.be/FLZvOKSCkxY</a></li></ul><h4>3.Deep Learning:</h4><p><strong>Deep learning</strong> (also known as <strong>deep structured learning</strong> or <strong>hierarchical learning</strong>) is part of a broader family of machine learning methods based on learning data representations, as opposed to task-specific algorithms. Learning can be supervised, semi-supervised or unsupervised.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/693/0*giROem57D1GCwTRN.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/690/0*Z0jQGVdPRGHNtkan.png" /></figure><p>Links for Deep Learning:</p><ul><li><a href="https://in.udacity.com/course/deep-learning--ud730">https://in.udacity.com/course/deep-learning--ud730</a></li><li><a href="http://deeplearning.net/">http://deeplearning.net/</a></li><li><a href="https://www.youtube.com/channel/UC9OeZkIwhzfv-_Cb7fCikLQ">https://www.youtube.com/channel/UC9OeZkIwhzfv-_Cb7fCikLQ</a></li><li><a href="https://www.nvidia.com/en-us/deep-learning-ai/education/">https://www.nvidia.com/en-us/deep-learning-ai/education/</a></li><li><a href="https://www.edx.org/course/deep-learning-explained-microsoft-dat236x-0">https://www.edx.org/course/deep-learning-explained-microsoft-dat236x-0</a></li></ul><h4>4.Robotics:</h4><p><strong>Artificial intelligence</strong> (AI) is arguably the most exciting field in robotics. It’s certainly the most controversial: Everybody agrees that a robot can work in an assembly line, but there’s no consensus on whether a robot can ever be intelligent.</p><p>Like the term “robot” itself, artificial intelligence is hard to define. Ultimate AI would be a recreation of the human thought process — a man-made machine with our intellectual abilities. This would include the ability to learn just about anything, the ability to reason, the ability to use language and the ability to formulate original ideas. Roboticists are nowhere near achieving this level of artificial intelligence, but they have made a lot of progress with more limited AI. Today’s AI machines can replicate some specific elements of intellectual ability.</p><p><a href="https://youtu.be/QVdQM47Av20">https://youtu.be/QVdQM47Av20</a></p><p>Crazy to learn about these Just Follow these links:</p><ul><li><a href="https://www.edx.org/course/robotics-columbiax-csmm-103x">https://www.edx.org/course/robotics-columbiax-csmm-103x</a></li><li><a href="https://courses.edx.org/courses/course-v1:PennX+ROBO1x+1T2017/course/">https://courses.edx.org/courses/course-v1:PennX+ROBO1x+1T2017/course/</a></li><li><a href="https://www.futurelearn.com/courses/begin-robotics">https://www.futurelearn.com/courses/begin-robotics</a></li><li><a href="http://roboticscourseware.org/">http://roboticscourseware.org/</a></li></ul><p>Thats it but there are many works of AI you can google it and check if you like .</p><h3>YOUTUBE CHANNELS TO FOLLOW:</h3><p>1)Siraj Raval — <a href="https://www.youtube.com/channel/UCWN3xxRkmTPmbKwht9FuE5A">https://www.youtube.com/channel/UCWN3xxRkmTPmbKwht9FuE5A</a></p><p>2)Sentdex — <a href="https://www.youtube.com/user/sentdex">https://www.youtube.com/user/sentdex</a></p><p>3)DeepLearning.TV — <a href="https://www.youtube.com/channel/UC9OeZkIwhzfv-_Cb7fCikLQ">https://www.youtube.com/channel/UC9OeZkIwhzfv-_Cb7fCikLQ</a></p><p>4)Google AI Adventures — <a href="https://www.youtube.com/playlist?list=PLIivdWyY5sqJxnwJhe3etaK7utrBiPBQ2">https://www.youtube.com/playlist?list=PLIivdWyY5sqJxnwJhe3etaK7utrBiPBQ2</a></p><p>5)Stanford University School of Engineering — <a href="https://www.youtube.com/user/stanfordeng/playlists">https://www.youtube.com/user/stanfordeng/playlists</a></p><h4>GOOGLE BLOGS:</h4><p><a href="https://cloud.google.com/blog/big-data/2017/01/learn-tensorflow-and-deep-learning-without-a-phd">https://cloud.google.com/blog/big-data/2017/01/learn-tensorflow-and-deep-learning-without-a-phd</a></p><h3>MOST USED FRAMEWORK FOR THE DEVELOPMENT OF AI:</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/352/0*8VcRYzn2489LWxqt.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/566/0*UUrYoFh706Hxmjvx.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/609/0*RW_AXCAK8mfcTo4M.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*lb2JeY5lN9iz-bwd." /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/609/0*icztfoPsYSmP7Eps.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/200/0*Pbs88QDYF9bFnSRv.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*rPseQsNfcZaXOFEN.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/179/0*-Ht1NdCHif6VTsTf.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/250/0*8kfx9j5pRxdDpyLD.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*5gfxjDa0fJI2Jl_4.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Y9BYEV8bKJAkjy5K.png" /></figure><h3>I AM AN AI DEVELOPER:</h3><p>After completely experienced with these resources and you are able to develop an AI application then woow congrats you are an AI Developer.These are just a beginner resources for your AI Career.Hope you like content in my blog.</p><p>Sharing is worth than anything.Kindly share this blog to your friends.</p><p>Feel free to comment your doubts or feedback about this blog in comment section below..</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/0*kQo2RZS3_zOrUdFd.jpg" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7ddffa2a586" width="1" height="1" alt=""><hr><p><a href="https://medium.com/buzzrobot/getting-started-with-ai-7ddffa2a586">Getting started with AI</a> was originally published in <a href="https://medium.com/buzzrobot">BuzzRobot</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>