Download What is AI PowerPoint Presentation


Login   OR  Register

Iframe embed code :

Presentation url :

Home / Business & Management / Business & Management Presentations / What is AI PowerPoint Presentation

What is AI PowerPoint Presentation

Ppt Presentation Embed Code   Zoom Ppt Presentation

PowerPoint is the world's most popular presentation software which can let you create professional What is AI powerpoint presentation easily and in no time. This helps you give your presentation on What is AI in a conference, a school lecture, a business proposal, in a webinar and business and professional representations.

The uploader spent his/her valuable time to create this What is AI powerpoint presentation slides, to share his/her useful content with the world. This ppt presentation uploaded by iamsimple in Business & Management ppt presentation category is available for free download,and can be used according to your industries like finance, marketing, education, health and many more.

About This Presentation

What is AI Presentation Transcript

Slide 1 - AI: Some (Biased) Highlights
Slide 2 - What is AI? Categories for definitions of AI Modern approach Classic View
Slide 3 - The Turing Test: Preliminaries Designed by Alan Turing (1950) The Turing test provides a satisfactory operational definition of AI It’s a behavioral test (i.e., test if a system acts like a human) Problem: it is difficult to make a mathematical analysis of it
Slide 4 - The Turing Test The Turing Test: a computer is programmed well enough to have a conversation with an interrogator (for example through a computer terminal) and passes the test if the interrogator cannot discern if there is a computer or a human at the other end
Slide 5 - The Turing Test vs. AI Fields For a program to pass the Turing Test, it needs to pass the exhibit the following capabilities: Natural language processing Knowledge representation Automated reasoning Machine learning
Slide 6 - Loebner Prize Each year (since 1994) a competition is made to see if a computer passes the Turing Test The first program to pass it will receive 100k Controversial: Minsky offer $100 if anyone finish it Still, it is interesting to observe capabilities Machines seems to have come close to fulfill Turing’s prediction (5 minutes)
Slide 7 - Other Predictions from Turing Predicted that by the year 2000 a computer will have 30% chances to fool a person for 5 minutes Anticipated the major argument against AI: The mathematical objection to AI
Slide 8 - The Mathematical Objection to AI: The Halting Problem Can we write a program in a C++, that recognizes if any program written in that language ends with a given input? Answer: No (Turing, 1940’s!) Mathematical Proof (CSE 318: Automata Theory): valid for any program running on yesterday’s, today's and tomorrow’s (foreseeable future) computers i = 2 While ( i ≠ c) i++ Does this program terminates for: c = 1? c = 10? No Yes
Slide 9 - Parenthesis: Halting Problem Proof by contradiction: assume that there is a program H that can determine if programs halt with a given input What would happen if we give H itself as input for H? We will obtain a paradox, similar to the one in the set: “Pedro shaves only the people who doesn’t shaves themselves” Does Pedro shaves himself?
Slide 10 - The Mathematical Objection to AI Argument against AI: a human can determine if a program ends or not Thus, computers are inferior as humans Argument against this argument: Are you sure that a human will be able to determine if a very complicated program will terminate with a certain input?
Slide 11 - AI: Genesis Logical reasoning calculus was conceived (Leibniz, 17 century) Leibiz’ motivation: solve intellectual arguments by calculation Boolean logic (Boole, 1847) Predicate Logic (Frege, 1879): Begriffsschrift Incompleteness Theorem (Goedel, 1940’s)
Slide 12 - AI: Some Historical Highlights Turing’s article about what machines can do Term AI is coined at the Dartmouth conference (1956) General Problem Solver (Newell & Simon; 1958) Period of great expectations
Slide 13 - Early Stages, Great Expectations (what they thought they could achieve) Jenna: What were you just thinking? Data: In that particular moment, I was reconfiguring the warp field parameters, analyzing the collected works of Charles Dickens, calculating the maximum pressure I could safely apply to your lips, considering a new food supplement for Spot... Jenna: I'm glad I was in there somewhere. (from the episode “In Theory” )
Slide 14 - AI: Some Historical Highlights (cont’d) Perceptrons: limits to neural networks (Minksy and Papert; 1969) Knowledge-based systems (1970’s) AI becomes an industry. Early successes of Expert systems
Slide 15 - AI: Some Historical Highlights (cont’d) It becomes clear that expert systems are hard to create (problem known as the Knowledge Acquisition bottle-neck) 1990’s: more consolidated approaches to AI, more realistic expectations, fielded applications: Program beats chess world champion Applications of machine learning to data-mining (CSE ???: Data Mining) Applications of Case-Based Reasoning to help-desk systems (CSE 395/495: Intelligent Decision Support Systems)
Slide 16 - Case-Based Reasoning: Definition A problem-solving methodology where solutions to similar, previous problems are reused to solve new problems.
Slide 17 - Problem-Solving with CBR Problem Space Solution Space p2 p1 s1 s2 p3 CBR(problem) = solution s4 s3
Slide 18 - CBR: First Example Example: Slide Creation - 12/1/03: talk@ cse15
Slide 19 - Other Applications of CBR Help-Desk Systems Help users when problem occurs Reduce overhead by lowering number of required support personnel. "Help you I can, yes, hmm?"
Slide 20 - What is Going on the Other Side Space of known problems for Tie Fighter Search is performed for a similar case
Slide 21 - What Attributes to consider for Similarity: ICM-Principle Independence: Attributes should represent independent features whenever possible Completeness: the attributes should be sufficient to determine if the case can be reused in a new situation Minimalist: The only attributes that should be included in a case are those used in to compute similarity (example: price paid for game is not relevant to identify problem running it)
Slide 22 - AI Planning Planning problem: Obtain a sequence of actions (plan) to achieve some goals in a particular situation C A B situation A B C goals move-C-from-A-to-Table move-B-from-Table-to-C move-A-from-Table-to-B plan
Slide 23 - Search Space A C B A B C A C B C B A B A C B A C B C A C A B A C B B C A A B C A B C A B C
Slide 24 - Hierarchical Task Network (HTN) Planning Principle: complex tasks are decomposed into simpler tasks. The goal is to decompose compound tasks into primitive tasks, which define actions changing the world. Fly(National, L.V. International) Travel(L.V. Int’nal,Lehigh) Travel(UMD,National) alternatives Travel from UMD to Lehigh University Travel by car Enough money for gasoline Roads are passable Seats available Travel by plane Enough money for air fare available Taxi(UMD,UMD-Metro) Metro(UMD-Metro,National) Taxi(L.V. Int’nal,Lehigh) Travel(UMD,National)
Slide 25 - Current Research: Controlling Unreal Tournament Bots
Slide 26 - BOTS Bot: software robot Receives information from the environment (game server) Represents knowledge about environment, beliefs, goals Acts according to knowledge (reason) Can learn from previous experiences
Slide 27 - Using HTNs to Represent Strategies A B C D E Capture the flag: Must bring the enemy’s flag to our camp Must not allow the enemy to take the flag 1 2 CTF Move(1,2,B) defend(1,B) capture(2,E)
Slide 28 - Use HTN Planning to Reason and Act A B C D E 1 2 CTF 1,2 Move(1,2,B) defend(1,B) 2 capture(2,E)
Slide 29 - How Can We Determine an Adequate Strategy? Use CBR!: View strategies and their situations (ICM-principle) as cases Map Own team Enemy … Define a similarity metric for comparing situations Encode algorithms to compute this similarity very fast (real time!)
Slide 30 - What Can the Bot Learn? New strategies: after every game store the strategy and make it available for future games Learn the similarity metric: some attributes are more important than others (for example, point B in the previous map seems crucial) Mistakes: some strategies maybe better than others
Slide 31 - But hasn’t Half-Life Done This? Half life: the best (?) first-person shooter with an amazing computer opponent (or “AI” as the game industry calls it) A military squad will act coordinately to attack you The behavior of the AI is hard-coded for the particular situation, you are always approaching from the same direction! What the computer industry calls “AI” is just a synonym for computer opponent. This can be hard-coded, or cheat, or use true AI techniques (Close Combat, Black and White)
Slide 32 - Current Status We developed an object-oriented architecture and connected it to the Unreal Tournament server (Todd Fisher) We are developing an HTN-based language for representing strategies (almost done) We are developing similarity assessments and algorithms for computing these quickly
Slide 33 - Conclusions AI is moving away from the idea of simulating humans and it is rather concentrating on developing programs that behave rationally But many of the techniques in use now are derived from the initial goal of simulating humans There is a broad range of deployed systems, companies using AI Key issues of AI: knowledge representation, reasoning, and machine learning Mathematical foundations are very important to understand the limitations of computing (CSE 318, CSE 340)