Time Table
------------
Introduction and Definitions: 2 weeks
Model abstraction ("learning" and "discovery"): 6 weeks
Model application ("doing"): 3 weeks
Integration of "learning" and "discovery" with "doing": 2 weeks
System analysis: 2 weeks
Course project: 1 week
Main References
---------------
Autonomous Learning from the Environment, Computer Science Press, 1994,
plus various papers on each of the techniques discussed in the class.
Lecture Plan
--------------
1. Aug 28 . Course organization (lecture schedule, homework description, final project), philosophy, (Why this topic?) history, (you decide) background (give a list). Definitions of Autonomous learning: actions, percepts, mental constructors, models, transparent and translucent environments, etc. Ch. 1,2 and related references.
2. Sep 2 . Tasks of autonomous learning (three tasks, learn;do;integrate), and views of other scientific fields, such as function approximation, adaptive control, cognitive psychology, and others. Ch. 3 and related references.
3. Sep 4. Model abstraction in transparent environment; Experiences spaces and model spaces, model construction via Direct Recording. Ch. 4.1-4.2.
4. Sep 9. Model abstraction in transparent environment; the problem of abstracting from attribute-based perceptions, and active learning of concepts. Ch. 4.3-4.4, and related references.
5. Sep 11. Model abstraction in transparent environment; the algorithms for attribute-based concept learning, Version Space, CDL1 and CDL2. Ch. 4.5 4.7-4.9.
6. Sep 16. Model abstraction in transparent environment; the problem of abstracting from structure or relation-based perceptions. Ch. 4.4, 4.10, and related references.
7. Sep 18. Model abstraction in transparent environment; algorithms. FOIL and CDL3. Ch. 4.6 and 4.10.
8. Sep 23. Model abstraction in transparent environment; Bayesian Probability, neural networks. Ch. 4.11-12 and related references.
9. Sep 25. Model abstraction using Bayesian belief networks. Related references.
10. Sep 30. Model abstraction in translucent environment; Problems and active learning of finite automata with reset. The L* algormithm. Ch. 5.1-5.2.
11. Oct 2. Model abstraction in translucent environment; Homing sequences and related algorithm. Ch. 5.3 and related references.
12. Oct 7. Model abstraction in translucent environment; active learning finite automata without reset, local distinguishing experiments. Ch. 5.4-5.7 and related references.
13. Oct 9. Model abstraction in translucent environment; discovering hidden features when learning prediction rules. Ch. 5.8 and related references.
14. Oct 14. Model abstraction in translucent environment; Stochastic automata. Ch. 5.9 and related references.
15. Oct 16. Model abstraction in translucent environment; Hidden Markov models. Ch. 5.9-5.10 and related references.
16. Oct 21. Model abstraction using prior knowledge, explanation-based learning. DeJong & Mooney, FOCL.
----------------------------------------------
17. Oct 23. Methods of model application. Types of planning.
18. Oct 28. Model application, searching for optimal solutions, Dynamic programming, A*, Q-learning. Ch 6.1 and related references.
19. Oct 30. Model application, searching for satisficing solutions, Real-time A*, Distal supervised learning, and symbolic goal regression. Ch 6.2 and related references.
20. Nov 4. Model application with Bayesian belief networks. Related references.
21. Nov 6. Model application, designing and learning from experimentation. Ch 6.3-6.5 and related references.
----------------------------------------------
22. Nov 11. Integration of model abstraction and model application in
transparent environment. Explanation-based reienforcement learning.
Ch. 7.1.1, 7.2.1, and related references.
23. Nov 13. Integration of model abstraction and model application in translucent environment. Ch. 7.1.2, 7.2.2, and related references.
24. Nov 18. Systems. LIVE architecture and details. Ch. 8, 9, 10, and related references.
25. Nov 20. Systems. LIVE performance (continues), Ch. 11,and related references.
26. Nov 25. Systems. systems for Map Learning. Ch. 12, and related references.
27. Dec 2. The future of autonomous learning systems, and project discussion (project assignment will be given on 11/12).
28. Dec 4. Invited lecture (floating)
Homework: At least three (TBA)
Final Project: