<?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 Eliz Ayça on Medium]]></title>
        <description><![CDATA[Stories by Eliz Ayça on Medium]]></description>
        <link>https://medium.com/@elizaycayilmaz?source=rss-d9ffd1754dca------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*tzIbrnalltdpSWBt_bxJdw.jpeg</url>
            <title>Stories by Eliz Ayça on Medium</title>
            <link>https://medium.com/@elizaycayilmaz?source=rss-d9ffd1754dca------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 23:08:22 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@elizaycayilmaz/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[SQL Server ile Veritabanı Programlama-Giriş]]></title>
            <link>https://elizaycayilmaz.medium.com/sql-server-ile-veritaban%C4%B1-programlama-giri%C5%9F-d4f170ab9bc6?source=rss-d9ffd1754dca------2</link>
            <guid isPermaLink="false">https://medium.com/p/d4f170ab9bc6</guid>
            <category><![CDATA[t-sql]]></category>
            <category><![CDATA[sql-server]]></category>
            <category><![CDATA[mssql]]></category>
            <category><![CDATA[database]]></category>
            <category><![CDATA[veritabanı]]></category>
            <dc:creator><![CDATA[Eliz Ayça]]></dc:creator>
            <pubDate>Wed, 20 Jan 2021 15:50:19 GMT</pubDate>
            <atom:updated>2021-01-20T15:58:04.642Z</atom:updated>
            <content:encoded><![CDATA[<p>Herkese, merhaba🤚Bilgisayar mühendisliğinden bu sene mezun oldum. Bu süreçte elimden geldiği kadar yeni şeyler öğrenmeye çalışıyorum.</p><p>Veri tabanı programcılığı çok geniş bir alan.(Hangi alan dar ki? 🤕)SQL ve T-SQL çok aranan bir özellik programcılar için sizde bu alanda gelişmek veya bu bilgileri edinmek istiyorsanız yazılarımı lütfen takip edin.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/367/1*2bjzCMZfCFRqpPwl8BqBvA.png" /></figure><blockquote>Genel Bakış</blockquote><p>Öncellikle bilmemiz gereken bazı temel şeyler var bunlara bakalım.</p><p><strong>Veri ve Bilgi bu kavramlar nelerdir farkı nelerdir?</strong></p><p><strong>Veri</strong> dediğimiz şey işlenmemiş ham enformasyondur. Örneğin; bir kişinin tansiyon ölçümleri bu veriler işlenmemiş analiz edilmemiş verilerdir.</p><p><strong>Bilgi</strong> dediğimiz şey ise anlamlandırılmış verilerdir.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/240/1*E0FNBfI75adUGPttwCyfMQ.png" /></figure><p><strong>Veri tabanı nedir?</strong></p><p><strong>Veri tabanı :</strong> ilişkili veya ilişkisiz verilerin belirli bir düzen içerisinde depolandığı özel yazılımlardır.</p><p>✍️Sorgulanabilir, değiştirilebilir veya silinebilirler.</p><p>✍️Disk üzerinde tutulurlar. Bu verilere sadece veri tabanı yazılımının SQL motoru tarafından erişilebilir.</p><p>✍️Dış sistemlere entegre amacıyla geliştirilen erişim araçlarıyla (driver)’da kullanılır.</p><p>✍️En temel kavramı connection’dur. Tüm yetkiler bu bağlantı üzerinden yönlendirilir.</p><p>Veri tabanı programlama da farklılıklar gösteriyor tabi ki .Şöyle diyelim bir çok SQL çeşidi var. Temelde hepsi ANSI SQL standardına uysa da kendilerine özgü tarafları olduğundan farklılaşmıştır.</p><p><strong>SQL Türleri, ne nerede kullanılır?</strong></p><p>Kurumsal sektörde çalışmayı düşünüyorsanız. Lisanslı ürünleri tercih etmelisiniz en çok kullanılanlar;</p><p>👉Microsoft- SQL/T-SQL</p><p>👉Oracle- PL/SQL👉</p><p>Açık kaynak kültürü olan bir ortamda çalışmak isteyenler(start up vs.);</p><p>👉PostgreSQL</p><p>👉MySQL</p><p>👉NoSQL</p><p>Biz bu eğitim serisinde SQL Server kullanacağız. SQL/T-SQL öğreneceğiz.</p><p>Bunun dışındakileri yukarıda belirttim kısaca. Daha detaylı araştırmasını yapabilirsiniz.</p><p>Takipte kalın 👩‍💻.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d4f170ab9bc6" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Searching Algorithms]]></title>
            <link>https://elizaycayilmaz.medium.com/searching-algorithms-79cc886f6e46?source=rss-d9ffd1754dca------2</link>
            <guid isPermaLink="false">https://medium.com/p/79cc886f6e46</guid>
            <category><![CDATA[searching-algorithm]]></category>
            <category><![CDATA[informed]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Eliz Ayça]]></dc:creator>
            <pubDate>Mon, 25 Nov 2019 00:42:11 GMT</pubDate>
            <atom:updated>2019-11-25T00:42:11.530Z</atom:updated>
            <content:encoded><![CDATA[<h4>Artificial Intelligence-5</h4><blockquote><strong>Measuring problem-solving problems</strong></blockquote><ul><li><strong>Completeness:</strong> Is the algorithm guaranteed to find a solution when there is one?</li><li><strong>Optimality:</strong> Does the strategy find the optimal solution?</li><li><strong>Time complexity:</strong> How long does it take to find a solution?</li><li><strong>Space complexity:</strong> How much memory is needed to perform the search?</li></ul><blockquote><strong>Uninformed Search Algorithms</strong></blockquote><p><em>While searching you have no clue whether one non-goal state is better than any other. Your search is blind. You don’t know if your current exploration is likely to be fruitful.</em></p><ul><li><strong>Breadth-First Search(BFS)</strong></li><li><strong>Uniform-Cost Search</strong></li><li><strong>Depth-First Search</strong></li><li><strong>Depth-Limited Search</strong></li><li><strong>Iterative Deepening Search</strong></li><li><strong>Bidirectional Search</strong></li></ul><h4><strong>Breadth-First Search</strong></h4><p>Expand shallowest unexpanded node</p><p>fringe is a<strong> first-in-first-out (FIFO)</strong> queue, i.e., new successors go at end of the queue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mU9SFi8YVHwdYgF3sOgKCg.png" /></figure><p><strong><em>Properties of Breadth First Search</em></strong></p><ul><li><strong><em>Complete?</em></strong> Yes it always reaches goal (if b is finite)</li><li><strong><em>Time? </em></strong>1+b+b 2+b 3+… +b d + (b d+1 -b)) = O(bd+1) (this is the number of nodes we generate)</li><li><strong><em>Space?</em></strong> O(b d+1 ) (keeps every node in memory, either in fringe or on a path to fringe).</li><li><strong><em>Optimal?</em></strong> Yes (if we guarantee that deeper solutions are less optimal, e.g. step-cost=1).</li></ul><h4><strong>Uniform-Cost Search</strong></h4><p>Expand node with smallest path cost g(n). <strong>“f(n)=g(n)”</strong></p><p>fringe = queue ordered by path cost Equivalent to breadth-first if all step costs all equal.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/464/1*sY96jwqbfpW6tp5oDBynSA.gif" /></figure><p><strong>S-A=1</strong>,S-B=5,S-C=15;<em>S-A-G=1+10=11&gt;5</em></p><p><strong>S-B=5,S-B-G=5+5=10&lt;11,15 <em>so it is our path.</em></strong></p><p><strong><em>Properties of Uniform-Cost Search</em></strong></p><ul><li><strong><em>Complete?</em></strong> Yes, if step cost ≥ ε (otherwise it can get stuck in infinite loops)</li><li><strong><em>Time? </em></strong># of nodes with path cost ≤ cost of optimal solution.</li><li><strong><em>Space?</em></strong> # of nodes with path cost ≤ cost of optimal solution.</li><li><strong><em>Optimal?</em></strong> Yes, for any step cost ≥ ε</li></ul><h4>Depth-First Search</h4><p>Expand deepest unexpanded node</p><p>fringe = Last In First Out (LIFO) queue, i.e., put successors at front</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tvuGLNrXVdO9qR1TSicVEQ.png" /></figure><p><strong><em>Properties of Depth-First Search</em></strong></p><ul><li><strong><em>Complete?</em></strong> No: fails in infinite-depth spaces Can modify to avoid repeated states along path</li><li><strong><em>Time? </em></strong>O(b m ) with m=maximum depth</li></ul><p><strong><em>-terrible if m is much larger than d</em></strong></p><p><strong><em>-but if solutions are dense, may be much faster than breadth-first</em></strong></p><ul><li><strong><em>Space?</em></strong> O(bm), i.e., linear space! (we only need to remember a single path +expanded unexplored nodes)</li><li><strong><em>Optimal?</em></strong> No (It may find a non-optimal goal first)</li></ul><h4>Depth-Limited Search</h4><p><em>The embarrassing failure of depth-first search in infinite state spaces can be alleviated by supplying depth-first search with a predetermined depth limit . That is, nodes at depth are treated as if they have no successors. This approach is called depth-limited search .Depth limit solves the</em><strong><em> infinite-path problem.</em></strong></p><h4>Iterative Deepening Search</h4><ul><li>To avoid the infinite depth problem of DFS, we can decide to only search until depth L, i.e. we don’t expand beyond depth L.</li></ul><p><strong><em>→ Depth-Limited Search</em></strong></p><ul><li>What if solution is deeper than L? → Increase L iteratively.</li></ul><p><strong><em>→ Iterative Deepening Search</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*K-24GduKvM_R8G0dMHKeqQ.png" /></figure><p><strong><em>Properties of Iterative Deepening Search</em></strong></p><ul><li><strong><em>Complete?</em></strong> Yes</li><li><strong><em>Time? </em></strong>O(b^d)</li><li><strong><em>Space?</em></strong> O(bd)</li><li><strong><em>Optimal?</em></strong> Yes, if step cost = 1 or increasing function of depth.</li></ul><h4>Bidirectional Search</h4><p><strong><em>Idea</em></strong></p><p>◼ simultaneously search forward from S and backwards from G</p><p>◼ stop when both “meet in the middle”</p><p>◼ need to keep track of the intersection of 2 open sets of nodes</p><p><strong><em>What does searching backwards from G mean</em></strong></p><p>◼ need a way to specify the predecessors of G</p><p>•this can be difficult,</p><p>• e.g., predecessors of checkmate in chess?</p><p>◼ which to take if there are multiple goal states?</p><p>◼ where to start if there is only a goal test, no explicit list?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/1*JjW0uoiZTBN0vcruOLoDAQ.png" /></figure><blockquote><strong><em>Summary of algorithms</em></strong></blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YQ6gdaTty6t4mALePUxfYw.png" /></figure><h4><strong>References:</strong></h4><ul><li>Prateek Joshi, “Artificial Intelligence with Python : A Comprehensive Guide to Building Intelligent Apps for Python Beginners and Developers”, 1 st Ed., Packt Publishing, 2017.</li><li>Stuart Russell and Peter Norvig, “Artificial Intelligence: A Modern Approach” , 3rd Ed., Prentice Hall, 2010.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=79cc886f6e46" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Solving Problems by Searching]]></title>
            <link>https://elizaycayilmaz.medium.com/solving-problems-by-searching-239abf47a832?source=rss-d9ffd1754dca------2</link>
            <guid isPermaLink="false">https://medium.com/p/239abf47a832</guid>
            <category><![CDATA[solving]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[searching-algorithm]]></category>
            <dc:creator><![CDATA[Eliz Ayça]]></dc:creator>
            <pubDate>Mon, 25 Nov 2019 00:41:19 GMT</pubDate>
            <atom:updated>2019-11-25T00:41:19.671Z</atom:updated>
            <content:encoded><![CDATA[<h4>Artificial Intelligence-4</h4><p><em>In which we see how an agent can find a sequence of actions that achieves its goals when no single action will do.</em></p><blockquote><strong>Why Search?</strong></blockquote><ul><li>To achieve goals or to maximize our utility we need to predict what the result of our actions in the future will be.</li><li>There are many sequences of actions, each with their own utility.</li><li>We want to find, or search for, the best one.</li></ul><h4><strong>Example:Romania</strong></h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*99pcDhBDlc9o305W2yEdOg.png" /></figure><ul><li>On holiday in Romania; currently in Arad.</li><li>Flight leaves tomorrow from Bucharest</li><li><strong>Formulate goal</strong>: be in Bucharest</li></ul><p><em>(first step of problem solving)</em></p><ul><li><strong>Formulate problem:</strong> -states: various cities -actions: drive between cities or choose next city</li></ul><p><em>( Problem Formulation is the process of deciding what actions and states to consider, given a goal.)</em></p><ul><li><strong>Find solution:</strong> sequence of cities, e.g., Arad, Sibiu, Fagaras, Bucharest</li></ul><p><strong><em>Now,we determine task environment of this example.</em></strong></p><p>◼ Static / Dynamic : Previous problem was <strong>static</strong>: no attention to changes in environment</p><p>◼ Observable / Partially Observable / Unobservable Previous : problem was <strong>observable</strong>: it knew its states at all times.</p><p>◼ Deterministic / Stochastic : Previous problem was <strong>deterministic</strong>: no new percepts were necessary, we can predict the future perfectly given our actions</p><p>◼ Discrete / Continuous: Previous problem was<strong> discrete</strong>: we can enumerate all possibilities</p><p>◼ Single Agent/Multi Agent:Previous problem was <strong>single agent</strong>:No other agents interacting with your cost function</p><p>◼ Sequential/Episodic:Previous problem was <strong>sequential</strong>:Decisions depend on past decisions</p><blockquote><strong>Problem Formulation</strong></blockquote><p>A problem is defined by four items:</p><ul><li><strong>initial state</strong> e.g., “at Arad“</li><li><strong>actions</strong> set of possible actions in current state x.</li><li><strong>transition model</strong> Result(x,a) = state that follows from applying action a in state x. e.g., Result(X2 (Arad), A4 (Arad → Zerind)) = X3 (Zerind)</li><li><strong>goal test</strong>, e.g., x = “at Bucharest”, Checkmate(x)</li><li><strong>path cost</strong> (additive) e.g., sum of distances, number of actions executed, etc. c(x,a,y) is the step cost, assumed to be ≥ 0</li><li>A <strong>solution </strong>is a sequence of actions leading from the initial state to a goal state.</li></ul><blockquote><strong>Toy Problems</strong></blockquote><p>Ex:Vacuum World</p><ul><li><strong>States:</strong> The state is determined by both the agent location and the dirt locations. The agent is in one of two locations, each of which might or might not contain dirt. Thus, there are 2 × 22 = 8 possible world states. A larger environment with n locations has n · 2n states.</li><li><strong>Initial state:</strong> Any state can be designated as the initial state.</li><li><strong>Actions: </strong>In this simple environment, each state has just three actions: Left, Right, and Suck. Larger environments might also include Up and Down.</li><li><strong>Transition model:</strong> The actions have their expected effects, except that moving Left in the leftmost square, moving Right in the rightmost square, and Sucking in a clean square have no effect.</li><li><strong>Goal test:</strong> This checks whether all the squares are clean.</li><li><strong>Path cost:</strong> Each step costs 1, so the path cost is the number of steps in the path.</li></ul><p>Ex:8-puzzle problem</p><ul><li><strong>States:</strong> A state description specifies the location of each of the eight tiles and the blank in one of the nine squares.</li><li><strong>Initial state:</strong> Any state can be designated as the initial state. Note that any given goal can be reached from exactly half of the possible initial states (Exercise 3.4).</li><li><strong>Actions:</strong> The simplest formulation defines the actions as movements of the blank space Left, Right, Up, or Down. Different subsets of these are possible depending on where the blank is.</li><li><strong>Transition model:</strong> Given a state and action, this returns the resulting state; for example, if we apply Left to the start state in Figure 3.4, the resulting state has the 5 and the blank switched.</li><li><strong>Goal test:</strong> This checks whether the state matches the goal configuration shown in Figure 3.4. (Other goal configurations are possible.)</li><li><strong>Path cost:</strong> Each step costs 1, so the path cost is the number of steps in the path.</li></ul><p>Ex:8-queens problem</p><ul><li><strong>States:</strong> Any arrangement of 0 to 8 queens on the board is a state.</li><li><strong>Initial state:</strong> No queens on the board.</li><li><strong>Actions:</strong> Add a queen to any empty square.</li><li><strong>Transition model:</strong> Returns the board with a queen added to the specified square.</li><li><strong>Goal test: </strong>8 queens are on the board, none attacked.</li></ul><h4>References:</h4><ul><li>Prateek Joshi, “Artificial Intelligence with Python : A Comprehensive Guide to Building Intelligent Apps for Python Beginners and Developers”, 1 st Ed., Packt Publishing, 2017.</li><li>Stuart Russell and Peter Norvig, “Artificial Intelligence: A Modern Approach” , 3rd Ed., Prentice Hall, 2010.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=239abf47a832" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Intelligent Agents]]></title>
            <link>https://elizaycayilmaz.medium.com/intelligent-agents-bbc10c5ebedd?source=rss-d9ffd1754dca------2</link>
            <guid isPermaLink="false">https://medium.com/p/bbc10c5ebedd</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Eliz Ayça]]></dc:creator>
            <pubDate>Sun, 24 Nov 2019 01:39:38 GMT</pubDate>
            <atom:updated>2019-11-24T01:48:22.218Z</atom:updated>
            <content:encoded><![CDATA[<h4>Artificial Intelligence-3</h4><blockquote><strong>Agents</strong></blockquote><p>An agent is anything that can be viewed as perceiving its environment through sensors and acting upon that environment through actuators.</p><p><strong>Human agent:</strong> eyes, ears, and other organs for sensors; hands, legs, mouth, and other body parts for actuators .</p><p><strong>Robotic agent:</strong> cameras and infrared range finders for sensors; various motors for actuators(effectors).</p><blockquote><strong>Agents and environments</strong></blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/334/1*b-1BrIIdpiOCh8rspe1tfA.png" /></figure><p>The agent function maps from percept histories to actions:</p><p><strong>[f: P* → A]</strong></p><p>The agent program runs on the physical architecture to produce f</p><p><strong>agent = architecture + program</strong></p><blockquote><strong>Vacuum-cleaner world</strong></blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/314/1*Voe5mcsKOIWsIu5oDgUwXw.png" /></figure><p><strong>Percepts:</strong> location and state of the environment,e.g., [A,Dirty], [B,Clean]</p><p><strong>Actions:</strong> Left, Right, Suck, NoOp</p><blockquote><strong>Rational agents</strong></blockquote><p><strong>Rational Agent:</strong> For each possible percept sequence, a rational agent should select an action that is expected to maximize its performance measure, based on the evidence provided by the percept sequence and whatever built-in knowledge the agent has.</p><p><strong>Performance measure:</strong> An objective criterion for success of an agent’s behavior</p><p><em>e.g., performance measure of a vacuum-cleaner agent could be amount of dirt cleaned up, amount of time taken, amount of electricity consumed, amount of noise generated, etc.</em></p><p><strong>Rationality</strong> is distinct from omniscience (all-knowing with infinite knowledge)</p><p>Agents can perform actions in order to modify future percepts so as to obtain useful information (information gathering, exploration)</p><p>An agent is autonomous if its behavior is determined by its own percepts &amp; experience (with ability to learn and adapt) without depending solely on build-in knowledge</p><blockquote><strong>Task Environment</strong></blockquote><p>Before we design an intelligent agent, we must specify its “task environment”: <strong>PEAS: </strong>Performance -Measure-Environment -Actuators -Sensors</p><ul><li><strong>Example: Agent = taxi driver</strong></li></ul><p><strong>— Performance measure:</strong> Safe, fast, legal, comfortable trip, maximize profits</p><p><strong>— Environment:</strong> Roads, other traffic, pedestrians, customers</p><p><strong>— Actuators: </strong>Steering wheel, accelerator, brake, signal, horn</p><p><strong>— Sensors:</strong> Cameras, sonar, speedometer, GPS, odometer, engine sensors, keyboard</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/412/1*X_mvysjGBg0o4ciC_4rk-Q.png" /></figure><blockquote><strong>Environment types</strong></blockquote><p><strong><em>Fully observable (vs. partially observable):</em></strong> An agent’s sensors give it access to the complete state of the environment at each point in time.</p><p><strong><em>Deterministic (vs. stochastic):</em></strong> The next state of the environment is completely determined by the current state and the action executed by the agent. (If the environment is deterministic except for the actions of other agents, then the environment is strategic)</p><p><strong><em>Episodic (vs. sequential):</em></strong> An agent’s action is divided into atomic episodes. Decisions do not depend on previous decisions/actions.</p><p><strong><em>Static (vs. dynamic): </em></strong>The environment is unchanged while an agent is deliberating. (The environment is semidynamic if the environment itself does not change with the passage of time but the agent’s performance score does)</p><p><strong><em>Discrete (vs. continuous):</em></strong> A limited number of distinct, clearly defined percepts and actions. How do we represent or abstract or model the world?</p><p><strong><em>Single agent (vs. multi-agent): </em></strong>An agent operating by itself in an environment. Does the other agent interfere with my performance measure?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/538/1*f84WWy9q46-8PiRXOBLZvQ.jpeg" /></figure><blockquote><strong>Agent types</strong></blockquote><p><strong>Five </strong>basic types in order of increasing generality:</p><ul><li>Table Driven agents</li><li>Simple reflex agents</li><li>Model-based reflex agents</li><li>Goal-based agents</li><li>Utility-based agents</li></ul><figure><img alt="Simple Reflex Agents" src="https://cdn-images-1.medium.com/max/298/1*GyzRuS_GnBROjIAPQRnSoA.png" /><figcaption>Simple Reflex Agent</figcaption></figure><p>The simplest kind of agent is the simple reflex agent. These agents select actions on the basis of the current percept, ignoring the rest of the percept history.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/282/1*jB_DUfsYIu4koqqvqKYs-A.png" /><figcaption>Model-Based Agent</figcaption></figure><p>The most effective way to handle partial observability is for the agent to keep track of the part of the world it can’t see now.</p><p>That is, the agent should maintain some sort of internal state that depends on the percept history and thereby reflects at least some of the unobserved aspects of the current state</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/283/1*OS1c9iVI0YRzFi3K2nrpAw.png" /><figcaption>Goal-Based Agent</figcaption></figure><p>Knowing something about the current state of the environment is not always enough to decide what to do.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/284/1*XZOB-3LR-fdJQCnyMluCow.png" /><figcaption>Utility-Based Agent</figcaption></figure><p>Goals alone are not enough to generate high-quality behavior in most environments.</p><h4>References:</h4><ul><li>Prateek Joshi, “Artificial Intelligence with Python : A Comprehensive Guide to Building Intelligent Apps for Python Beginners and Developers”, 1 st Ed., Packt Publishing, 2017.</li><li>Stuart Russell and Peter Norvig, “Artificial Intelligence: A Modern Approach” , 3rd Ed., Prentice Hall, 2010.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bbc10c5ebedd" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Definitions of AI]]></title>
            <link>https://elizaycayilmaz.medium.com/definitions-of-ai-5efb8989db09?source=rss-d9ffd1754dca------2</link>
            <guid isPermaLink="false">https://medium.com/p/5efb8989db09</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Eliz Ayça]]></dc:creator>
            <pubDate>Sun, 24 Nov 2019 01:04:48 GMT</pubDate>
            <atom:updated>2019-11-24T01:06:15.174Z</atom:updated>
            <content:encoded><![CDATA[<h4>Artificial Intelligence-2</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/850/1*tlJgGNhLDR_3awzVIWe-aw.png" /></figure><h4><strong>Acting humanly:” The Turing Test approach”</strong></h4><ul><li>The legendary computer scientist and mathematician, Alan Turing, proposed the Turing Test to provide a definition of intelligence. It is a test to see if a computer can learn to mimic human behavior. He defined intelligent behavior as the ability to achieve human-level intelligence during a conversation. This performance should be sufficient to trick an interrogator into thinking that the answers are coming from a human.</li><li>To see if a machine can do this, he proposed a test setup: he proposed that a human should interrogate the machine through a text interface. Another constraint is that the human cannot know who’s on the other side of the interrogation, which means it can either be a machine or a human. To enable this setup, a human will be interacting with two entities through a text interface. These two entities are called respondents. One of them will be a human and the other one will be the machine.</li><li>The respondent machine passes the test if the interrogator is unable to tell whether the answers are coming from a machine or a human. The following diagram shows the setup of a Turing Test:</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/257/1*jrTDnXp4aVju2aoxGiI8hg.png" /></figure><p>As you can imagine, this is quite a difficult task for the respondent machine. There are a lot of things going on during a conversation. At the very minimum, the machine needs to be well versed with the following things:</p><ul><li><strong>Natural Language Processing:</strong> The machine needs this to communicate with the interrogator. The machine needs to parse the sentence, extract the context, and give an appropriate answer.</li><li><strong>Knowledge Representation:</strong> The machine needs to store the information provided before the interrogation. It also needs to keep track of the information being provided during the conversation so that it can respond appropriately if it comes up again.</li><li><strong>Reasoning:</strong> It’s important for the machine to understand how to interpret the information that gets stored. Humans tend to do this automatically to draw conclusions in real time.</li><li><strong>Machine Learning:</strong> This is needed so that the machine can adapt to new conditions in real time. The machine needs to analyze and detect patterns so that it can draw inferences.</li></ul><h4>Thinking humanly: The cognitive modeling approach</h4><p>Once we gather enough data, we can create a model to simulate the human process. This model can be used to create software that can think like humans. Of course this is easier said than done! All we care about is the output of the program given a particular input. If the program behaves in a way that matches human behavior, then we can say that humans have a similar thinking mechanism.</p><p>Within computer science, there is a field of study called <strong>Cognitive Modeling </strong>that deals with simulating the human thinking process. It tries to understand how humans solve problems. It takes the mental processes that go into this problem solving process and turns it into a software model. This model can then be used to simulate human behavior. Cognitive modeling is used in a variety of AI applications such as deep learning, expert systems, Natural Language Processing, robotics, and so on.</p><p>The interdisciplinary field of cognitive science brings together computer models from AI and experimental techniques from psychology to construct precise and testable theories of the human mind.</p><h4>Thinking rationally: The “laws of thought” approach</h4><p>Rationality refers to doing the right thing in a given circumstance.</p><p>The Greek philosopher Aristotle was one of the first to attempt to codify “right thinking,” that is, irrefutable reasoning processes. His <strong>syllogisms</strong> provided patterns for argument structures that always yielded correct conclusions when given correct premises — for example, <em>“Socrates is a man; all men are mortal; therefore, Socrates is mortal.” </em>These laws of thought were supposed to govern the operation of the mind; their study initiated the field called <strong>logic.</strong></p><blockquote><strong>There are two main obstacles to this approach.</strong></blockquote><blockquote><strong><em>1.</em></strong><em> It is not easy to take informal knowledge and state it in the formal terms required by logical notation, particularly when the knowledge is less than 100% certain.</em></blockquote><blockquote><strong><em>2. </em></strong><em>There is a big difference between solving a problem “in principle” and solving it in practice. Even problems with just a few hundred facts can exhaust the computational resources of any computer unless it has some guidance as to which reasoning steps to try first.</em></blockquote><h4>Acting rationally: The rational agent approach</h4><p>An agent is just something that acts (agent comes from the Latin agere, to do). Of course, all computer programs do something, but computer agents are expected to do more: operate autonomously, perceive their environment, persist over a prolonged time period, adapt to change, and create and pursue goals.</p><p>Rational agents need to be performed in such a way that there is maximum benefit to the entity performing the action. An agent is said to act rationally if, given a set of rules, it takes actions to achieve its goals. It just perceives and acts according to the information that’s available. This system is used a lot in AI to design robots when they are sent to navigate unknown terrains.</p><p>All the skills needed for the Turing Test also allow an agent to act rationally. Knowledge representation and reasoning enable agents to reach good decisions. We need to be able to generate comprehensible sentences in natural language to get by in a complex society.</p><blockquote><strong>The rational-agent approach has two advantages over the other approaches.</strong></blockquote><blockquote><strong><em>1.</em></strong><em> It is more general than the “laws of thought” approach because correct inference is just one of several possible mechanisms for achieving rationality.</em></blockquote><blockquote><strong><em>2.</em></strong><em> It is more amenable to scientific development than are approaches based on human behavior or human thought.</em></blockquote><h4>References:</h4><p>•Prateek Joshi, “Artificial Intelligence with Python : A Comprehensive Guide to Building Intelligent Apps for Python Beginners and Developers”, 1 st Ed., Packt Publishing, 2017.</p><p>•Stuart Russell and Peter Norvig, “Artificial Intelligence: A Modern Approach” , 3rd Ed., Prentice Hall, 2010.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5efb8989db09" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What is Artificial Intelligence?]]></title>
            <link>https://elizaycayilmaz.medium.com/what-is-artificial-intelligence-31facb336e0d?source=rss-d9ffd1754dca------2</link>
            <guid isPermaLink="false">https://medium.com/p/31facb336e0d</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Eliz Ayça]]></dc:creator>
            <pubDate>Sun, 24 Nov 2019 00:38:23 GMT</pubDate>
            <atom:updated>2019-11-24T00:51:18.235Z</atom:updated>
            <content:encoded><![CDATA[<h4>Artificial Intelligence-1</h4><p><strong><em>Artificial Intelligence(AI)</em></strong><em> is a way to make machines </em><strong><em>think and behave </em></strong><em>intelligently.These machines are controlled by software inside them,so AI has a lot to do with intelligent software programs that control these machines.It is a science of finding theories and methodologies that can help machines understand the world and accordingly react to situations in the same way that humans do.</em></p><p>AI is closely related to the <strong>study of human brain</strong>.Researchers believe that AI can be accomplished by understanding how the human brain works.By mimicking the way the human brain learns,thinks,and takes action,we can build a machine that can do the same.This can be used as a platform to develop intelligent systems that are capable of learning.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/275/1*15VYrKsogCnCi73S2aJoIA.jpeg" /></figure><h4>Why do we need to study AI?</h4><p>Even though the human brain is great at analyzing things around us,it cannot keep up with the preceding conditions.Hence,we need to design and develop intelligent machines that can do this.We need AI systems that can:</p><ul><li>Handle large amounts of data in an efficient way.With the advent of Cloud Computing,we are now able to store huge amounts of data.</li><li>Ingest data simultaneously from multiple sources without any lag.</li><li>Index and organize data in a way that allows us to derive insights.</li><li>Learn from new data and update constantly using the right learnin algorithms.</li><li>Think and respond to situations based on the conditions in real time.</li></ul><p>AI techniques are actively being used to make existing machines smarter,so that they can execute faster and more efficiently.</p><h4><strong>Applications of AI:</strong></h4><ul><li><strong>Computer Vision:</strong> These are the systems that deal with visual data such as images and videos. These systems understand the content and extract insights based on the use case. For example, Google uses reverse image search to search for visually similar images across the Web.</li><li><strong>Natural Language Processing:</strong> This field deals with understanding text. We can interact with a machine by typing natural language sentences. Search engines use this extensively to deliver the right search results.</li><li><strong>Speech Recognition:</strong> These systems are capable of hearing and understanding spoken words. For example, there are intelligent personal assistants on our smartphones that can understand what we are saying and give relevant information or perform an action based on that.</li><li><strong>Expert Systems:</strong> These systems use AI techniques to provide advice or make decisions. They usually use databases of expert knowledge areas such as finance, medicine, marketing, and so on to give advice about what to do next.</li><li><strong>Games:</strong> AI is used extensively in the gaming industry. It is used to design intelligent agents that can compete with humans. For example, AlphaGo is a computer program that can play the strategy game Go. It is also used in designing many other types of games where we expect the computer to behave intelligently.</li><li><strong>Robotics:</strong>Robotic systems actually combine many concepts in AI. These systems are able to perform many different tasks. Depending on the situation, robots have sensors and actuators that can do different things. These sensors can see things in front of them and measure the temperature, heat, movements, and so on. They have processors on board that compute various things in real time. They are also capable of adapting to the new environments.</li></ul><h4>Branches of AI:</h4><ul><li><strong>Machine Learning and pattern recognition:</strong> : This is perhaps the most popular form of AI out there. We design and develop software that can learn from data. Based on these learning models, we perform predictions on unknown data. One of the main constraints here is that these programs are limited to the power of the data. If the dataset is small, then the learning models would be limited as well.</li><li><strong>Logic-based AI:</strong> Mathematical logic is used to execute computer programs in logic-based AI. A program written in logic-based AI is basically a set of statements in logical form that express facts and rules about a particular problem domain. This is used extensively in pattern matching, language parsing, semantic analysis, and so on.</li><li><strong>Knowledge representation:</strong> The facts about the world around us need to be represented in some way for a system to make sense of them. The languages of mathematical logic are frequently used here. If knowledge is represented efficiently, systems can be smarter and more intelligent. Ontology is a closely related field of study that deals with the kinds of objects that exist. It is a formal definition of the properties and relationships of the entities that exist in a particular domain. This is usually done with a particular taxonomy or a hierarchical structure of some kind. The following diagram shows the difference between information and knowledge:</li><li><strong>Planning:</strong> This field deals with optimal planning that gives us maximum returns with minimal costs. These software programs start with facts about the particular situation and a statement of a goal. These programs are also aware of the facts of the world, so that they know what the rules are. From this information, they generate the most optimal plan to achieve the goal.</li><li><strong>Heuristics:</strong> A heuristic is a technique used to solve a given problem that’s practical and useful in solving the problem in the short term, but not guaranteed to be optimal. This is more like an educated guess on what approach we should take to solve a problem. In AI, we frequently encounter situations where we cannot check every single possibility to pick the best option. So we need to use heuristics to achieve the goal. They are used extensively in AI in fields such as robotics, search engines, and so on.</li><li><strong>Genetic Programming:</strong> Genetic programming is a way to get programs to solve a task, by mating programs and selecting the fittest. The programs are encoded as a set of genes, using an algorithm to get a program that is able to perform the given task really well.</li><li><strong>Search:</strong> The Search techniques are used extensively in AI programs. These programs examine a large number of possibilities and then pick the most optimal path. For example, this is used a lot in strategy games such as Chess, networking, resource allocation, scheduling, and so on.</li></ul><h4>References:</h4><ul><li>Prateek Joshi, “Artificial Intelligence with Python : A Comprehensive Guide to Building Intelligent Apps for Python Beginners and Developers”, 1 st Ed., Packt Publishing, 2017.</li><li>•Stuart Russell and Peter Norvig, “Artificial Intelligence: A Modern Approach” , 3rd Ed., Prentice Hall, 2010.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=31facb336e0d" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>