Presentation is loading. Please wait.

Presentation is loading. Please wait.

Introduction to AI & AI Principles (Semester 1) REVISION WEEK 1 (2008/09) John Barnden Professor of Artificial Intelligence School of Computer Science.

Similar presentations


Presentation on theme: "Introduction to AI & AI Principles (Semester 1) REVISION WEEK 1 (2008/09) John Barnden Professor of Artificial Intelligence School of Computer Science."— Presentation transcript:

1 Introduction to AI & AI Principles (Semester 1) REVISION WEEK 1 (2008/09) John Barnden Professor of Artificial Intelligence School of Computer Science University of Birmingham, UK

2 TODAY (Tuesday) uNature of exam (refining the info given in Week 11) uReview of material (extending the review in Week 11) uQuestions uNB Office Hours: l Friday 1 May, 3:00-4:00 l Friday 8 May, 4:00-5:00

3 Nature of the Examination

4 Format of AI Principles Exam uThree hours long. uFirst half (1.5 hours): almost exactly the same as the Intro-AI exam (see next slide). uSecond half: to be explained by Dean Petters in Revision Week 2. uUse of material: l First half can be done on the basis entirely of my material. l Second half can be done on the basis entirely of Dean’s material. l But you’re free to use his material in my half or mine in his half as appropriate.

5 Format of AI Intro Exam uOne and a half hours long. uDo 5 out of 6 questions. uMost question parts: broadly similar in style to exercises you did during Semester 1. uOne question is essay-like and allows considerable latitude as to what aspects of AI you address and what material you bring to bear. uThe rest are mostly on specific technical things, with a couple of free-wheeling aspects here and there.

6 AI-Intro Material uMy own lecture material, with some exclusions (see Week 11 part of Slides page) uAnswers / additional notes for Exercises. uAndrea Arcuri’s lecture on learning, with some exclusions. uBullinaria slides (again with some exclusions): l Semantic Networks (and my own notes on these slides) l Production Systems (and my own notes on these slides) l Expert Systems uTextbook chapters (or chapter parts) in the Weekly Reading Assignments on module webpage.

7 AI-Intro Material, contd uDon't be spooked by previous examinations, especially those from before 06-07!! There have been a lot of changes. Also, quite a few since last year. uKnowledge of textbook chapters or chapter parts other than those I've listed ISN’T expected. uKnowledge of Bullinaria slides other than those I point to from my list of weekly lecture slides ISN’T expected. uKnowledge of fine technical details in book chapters ISN’T expected. (I’m only expecting the main concepts and overall grasp of main examples.) uBut of course knowledge of all the above types could be helpful and impressive.

8 REVIEW of the material

9 Main Topics Covered u Representation and reasoning, in l logic l production systems l semantic networks. What we need to represent: entities (incl. situations, feelings, …), properties, relationships, propositional structure, quantification, … uPlanning (a type of reasoning). uSearch. uNatural Language difficulties as illustration of why AI is difficult. uKnowledge and reasoning needed in natural language understanding and operating in practical scenarios such as Hot Drinks and Shopping Trip. uLearning.

10 Main Detailed Techniques uExpressing information in logic. uExpressing information in semantic networks. uApplying production system rules (forwards or backwards, but fine detail only expected in forwards case). uDoing simple logical proofs. uSearch (fine detail not expected for best-first and A*). l Search as applied to route-finding. l Search as applied to planning delivery of a drink.

11 General Themes in AI uWhy everyday AI is difficult. l Language processing, vision, planning, common-sense reasoning, etc. u“Intelligence” and its connection to “stupidity”. l What looks like stupidity is often the understandably-incorrect application of efficient heuristics (rules of thumb) without which we and our AI cousins would be in a mess. uContd. ……

12 General Themes in AI, contd. uUncertain, vague, conflicting, missing, or diverse info. Huge amounts of info, of varying relevance. uHence: search, satisficing, graceful degradation, heuristics. uContext-sensitivity; incl. relativity to agents’ purposes (e.g., in vision and language interpretation). uTask variability, learning, adaptation, repair (e.g., of plans). uDeclarative/procedural trade-off. uGoal-directedness (backwards chaining) in reasoning and search.

13 A General Theme in AI uUncertain, vague, conflicting, missing, diverse, extensive info: l Amply shown by Hot Drinks, Shopping Trip and Crime scenarios, and by natural language examples. l Use of default rules and conflict resolution in PSs l Use of defaults and exceptions in SNs. l Contributes to need for search. Non-optimality (satisficing) in (some) search. l Use of heuristics in search. l Need for learning. l Graceful degradation in (e.g.) neural networks.

14 A General Theme in AI uSearch: l In planning (incl. route-finding, game-playing, …) l In deduction l In operation of Production Systems l In reasoning in Semantic Networks l In learning, particularly genetic algorithms automatically finding good weights for a neural network

15 General Theme: Heuristics uPS rules that leave out details and complications, and that are at best DEFAULTS uConflict resolution methods in PSs. uThe information attached to actions in planning about what changes (or doesn’t change) is typically defeasible. l On what doesn’t change: see the Planning 1 chapter in Callan about the important frame problem. uIn search in general: l Pruning l Action ordering in depth-first search l Evaluation functions in best-first search, incl. Heuristic functions in A* search. l Choice of search strategy, incl. backwards vs. forwards.

16 Rough Sequence of Topics uIntroduction: l what AI is l why we do it l how it differs from ordinary CS l application areas l expert versus everyday AI.

17 Topic Sequence contd: Challenge of AI uIntroductory examples from language. l CAUTION CHILDREN l “John got to his front door but realized he didn’t have his key.” l Context-sensitivity of language; knowledge and reasoning needed. uKnowledge and reasoning needed in Hot Drinks, Shopping Trip and Crime scenarios. l Knowledge variety, uncertainty, vagueness, missing info, … uVision and movement. l Context-sensitivity, purpose-sensitivity, ambiguity, …

18 Sequence of Topics, contd. uDetailed planning of delivery of one drink. l Search, forwards versus backwards chaining, goal- directedness l Knowledge needed about preconditions and (non-)effects of actions uSearch: general nature, example applications. uIntroduction to logic representation. uReasoning about a static situation using Production Systems. l CAUTION: different from planning, = reasoning about moving between different possible situations.

19 Sequence of Topics, contd. uMore on natural language difficulties. l vagueness l quantification subtleties l context-sensitivity l syntactic ambiguity, incl. PP attachment l some advanced topics: speech acts, mental states, metonymy, metaphor

20 Sequence of Topics, contd. uSearch detail (in route finding for Shopping Trip) l Different search strategies: depth-first, breadth-first, best-first, A* l Optimality or otherwise l Ordering and pruning heuristics l Evaluation/heuristic functions.

21 Sequence of Topics, contd. uMore on logic representation. uLogical deduction. l Inference rules in deduction versus production systems l Soundness l Fiddling around needed in deduction l Reduction of fiddling around by using Resolution l Reasoning by contradiction l Declarative/procedural trade-off. uLogical deduction versus using production systems. uReasoning as search.

22 Sequence of Topics, contd. uRepresentation and reasoning in Semantic Networks l Localization of info at nodes l Different types of link l Taxonomy (instances and subtypes) l Defaults and exceptions l Intersection search uMore on Production Systems. l Rule instantiations l Conflict resolution uExpert Systems

23 Sequence of Topics, contd. uLearning (Andrea Arcuri lecture in Week 9). l General characteristics l Neural Networks l Evolutionary Computation and Genetic Programming l Naïve Bayes Classifiers (not expected for exam)

24 Questions


Download ppt "Introduction to AI & AI Principles (Semester 1) REVISION WEEK 1 (2008/09) John Barnden Professor of Artificial Intelligence School of Computer Science."

Similar presentations


Ads by Google