Tuesdays, 11:00 a.m. – 12:30 p.m.
 Winter Quarter, 2013
Higher Education Center (HEC), Medford
Room 226

Instructor: Nils Nilsson

Course Web Page:

Special Note: The class is now over. No class on March 12.

Course Content

Overview: What is Intelligence? 
Intelligence is that quality that enables an entity to function appropriately and with foresight in its environment. Intelligent behavior can be arrayed along a continuum.

Plan of the course:  A continuum of ever-more capable "agents"

Prologue to the course: Examples of intelligent agents

Part I:  Reactive Agents
    Agents that react to simple inputs
        Topics: Rule-based actions, several examples
    Making agents more intelligent by increasing their perceptual abilities
        Topics: visual perception, multi-sensor integration, neural network learning, deep hierarchical networks, models of the cortex
    Making agents more intelligent by giving them "memory"
        Topics: maps and map learning, states, how to represent states, state tables, state-action
    Learning what action to take depending on state
        Topics: Reinforcement learning, model helicopter control, TDgammon, summary
Part II:  Agents that Make Plans
    What is required for a planning agent
        Topics: State representation, how actions change state, how to recognize a goal state
    Representing states and actions in a graph
        Topics: Finding paths in maps, the 8-puzzle, breadth-first search, heuristic search (A*)    
    Representing states by lists of what's true and how these lists change due to actions
        Topics:  STRIPS rules, forward and backward search
    Games (Adversarial Search)
         Topics:  Limits on search, evaluation functions, minimax search, some commercial chess and Go

Part III:  Agents that Reason
    Logical reasoning
        Topics: syllogisms, logic, deductions vs abductions, proving theorems, applications of automated theorem proving
    Semantic networks and taxonomies
CYC, Word Net, Image Net
    Using uncertain information
        Topics:  Probabilities, Bayesian networks

Part IV:  Agents that Understand Human Language
Speech recognition
        Topics:  waveform features -> phonemes -> words, hidden markov models

    Text understanding
        Topics: syntactic, semantic, and pragmatic analyses
    Use of Large Corpora

Part V:  Evolutionary AI
    Topics: Genetic Algorithms, Genetic Programming

Part VI:  Putting it all Together -- Robotics

    Topics: Examples of robots, robot architectures

Part VII:  Social Implications of AI
    Topics:  Benificial and Worrisome Effects, Military Robots, Privacy, Employment, . . . .

Links to  lecture slides

(Movies and animations do not play in the pdf files, but they should work in the ppsx files.)

Jan 8 PDF Format (30 MB)  
Jan 8 PPSX Format (24 MB)

Jan 15 PDF Format (9 MB)  
Jan 15 PPSX Format (10 MB)

Jan 22 PDF Format (10.6 MB)  
Jan 22 PPSX Format (9.6 MB)

Jan 29 PDF Format (5.3 MB) 
Jan 29 PPSX Format (7.3 MB)  

Feb 5 PDF Format (3.3 MB)  
Feb 5 PPSX Format (35 MB) 

Feb 12 PDF Format (4.7 MB)
Feb 12 PPSX Format (3.4 MB)

Feb 19 PDF Format (4.1 MB)  
Feb 19 PPSX Format (45 MB) -- This link actually takes you to my Stanford site

Feb 26 PDF Format (24 MB)  
Feb 26 PPSX Format (300 MB !)

Mar 5 PDF Format (1.1 MB)  
Mar 5 PPSX Format (1.1 MB)