Presentation is loading. Please wait.

Presentation is loading. Please wait.

Experience-Oriented Artificial Intelligence Rich Sutton with special thanks to Michael Littman, Doina Precup, Satinder Singh, David McAllester, Peter Stone,

Similar presentations


Presentation on theme: "Experience-Oriented Artificial Intelligence Rich Sutton with special thanks to Michael Littman, Doina Precup, Satinder Singh, David McAllester, Peter Stone,"— Presentation transcript:

1 Experience-Oriented Artificial Intelligence Rich Sutton with special thanks to Michael Littman, Doina Precup, Satinder Singh, David McAllester, Peter Stone, Lawrence Saul, and Harry Browne

2 Experience matters! Not in the obvious sense - that you have to do a thing many times to get good at it But just in the sense that you do things, that you live a life that you take actions, receive sensations that you pass through a trajectory of states over time This is so obvious that it passes unnoticed Like air, gravity

3 the actions taken and the sensations received, by the agent from its world a continuing time sequence over the life of the agent Experience is the minimal ontology Experience is AgentWorld experience

4 Experience matters, and must be respected Experience matters because It is what life is all about. Experience is the final common path, the only result of all that goes on in the agent and world

5 Experience is the most prominant feature of the computational problem we call AI It’s the central data structure, revealed and chosen over time It has a definite temporal structure  Order is important  Speed of decision is important There is a continuous flow of long duration (a lifetime!)  not a sequence of isolated interactions, whose order is irrelevant Experience matters computationally

6 Experience in AI Many, many AI systems have no experience They don't have a life! Expert Systems Knowledge bases like CYC Question-answering systems Puzzle solvers, or any planner that is designed to receive problem descriptions and emit solutions Part of the new popularity of agent-oriented AI is that it highlights experience Other AI systems have experience, but don’t “respect” it

7 Orienting around experience suggests radical changes in AI Knowledge of the world should be knowledge of possible experiences Planning should be about foreseeing and controlling experience The state of the world should be a summary of past experience, relevant to future experience Yet we rarely see these basic AI issues discussed in terms of experience Is it possible or plausible that they could be? Yes! Would it matter if they were? Yes!

8 I am not claiming that knowledge comes from experience. (I take no position on the nature/nuture controvery) But only that knowledge is about experience. And that, given that, it should be predictive.

9 Key Points Computational Theory vs. just making it work  What to compute and why  Experience is central to AI  Knowledge should be about experience The minimal ontology Grounding in experience from the bottom up A computational theory of knowledge must support  Abstraction  Composition  Decomposition - Explicitness, verifiability Such Modularity is the whole point of knowledge

10 Outline Experience as central to AI Predictive knowledge in General Generalized Transition Predictions (GTPs, or option models) Planning with GTPs (rooms-world example) State as predictions (PSRs) Prospects and conclusion

11 The I/O View of the World We are used to taking an I/O view of the mind, of the agent It does not matter what it is physically made of What matters is what it does So we should be willing to consider the same I/O view of the world It does not matter what it is physically made of What matters is what it does The only thing that matters about the world is the experience it generates

12 Then the only thing to know or say about the world is what experience it generates Thus, world knowledge must really be about future experience. In other words, it must be a prediction

13 AI could be about Predictions Hypothesis: Knowledge is predictive About what-leads-to-what, under what ways of behaving What will I see if I go around the corner? Objects: What will I see if I turn this over? Active vision: What will I see if I look at my hand? Value functions: What is the most reward I know how to get? Such knowledge is learnable, chainable, verifiable Hypothesis: Mental activity is working with predictions Learning them Combining them to produce new predictions (reasoning) Converting them to action (planning, reinforcement learning) Figuring out which are most useful

14 Philosophical and Psychological Roots Like classical british empiricism (1650–1800)  Knowledge is about experience  Experience is central But not anti-nativist (evolutionary experience) Emphasizing sequential rather than simultaneous events  Replace association/contiguity with prediction/contingency Close to Tolman’s “Expectancy Theory” (1932–1950)  Cognitive maps, vicarious trial and error Psychology struggled to make it a science (1890–1950)  Introspection  Behaviorism, operational definitions  Objectivity

15 Tolman & Honzik, 1930 “Reasoning in Rats” Food box Path 1 Path 3 Path 2 Block B Block A Start box

16 An old, simple, appealing idea Mind as prediction engine! Predictions are learnable, combinable They represent cause and effect, and can be pieced together to yield plans Perhaps this old idea is essentially correct. Just needs  Development, revitalization in modern forms  Greater precision, formalization, mathematics  The computational perspective to make it respectable  Imagination, determination, patience –Not rushing to performance

17 Outline Experience as central to AI Predictive knowledge in general Generalized Transition Predictions (GTPs, or option models) Planning with GTPs (rooms-world example) State as predictions (PSRs) Prospects and conclusion

18 In steps of increasing expressiveness:  Simple state-transition predictions  Mixtures of predictions  Closed-loop termination  Closed-loop action conditioning Machinery for General Transition Predictions

19 Experience 1-step Prediction stateaction AB a k-step Prediction AB  The Simplest Transition Predictions

20 Mixtures of k-step Predictions: Terminating over a period of time Where will I be in 10–20 steps? Where will I be in roughly k steps? now k=10 steps k=20 steps k steps Arbitrary termination profiles are possible short term medium term long term But sometimes anything like this is too loose and sloppy... now time steps of interest

21 Closed-loop Termination Terminate depending on what happens E.g., instead of “Will I finish this report soon” which uses a soft termination profile: Use “Will I be done when my boss gets here?” 1 hr probably in about an hour Prob. time 1 0 Prob. boss arrives only one precise but uncertain time matters

22 Closed-loop termination allows time specification to be both flexible and precise Instead of “what will I see at t +100?” Can say “what will I see when I open the box?” Will we elect a black or a woman president first? Where will the tennis ball be when it reaches me? What time will it be when the talk starts? or “when John arrives?” “when the bus comes?” “when I get to the store?” A substantial increase in expressiveness

23 Closed-loop Action Conditioning Each prediction has a closed-loop policy Policy: States --> Actions (or Probs.) If you follow the policy, then you predict and verify  Otherwise not  If partly followed, temporal-difference methods can be used

24 General Transition Predictions (GTPs) Closed-loop terminations and policies Correspond to arbitrary experiments and the results of those experiments What will I see if I go into the next room? What time will it be when the talk is over? Is there a dollar in the wallet in my pocket? Where is my car parked? Can I throw the ball into the basket? Is this a chair situation? What will I see if I turn this object around?

25 Anatomy of a General Transition Prediction 1 Predictor Recognizes the conditions, makes the prediction 2 Experiment - policy - termination condition - measurement function(s) knowledge verifier States Measurement space Actions

26 Room-to-Room GTPs (General Transition Predictions) up down rightleft (to each room's 2 hallways) Fail 33% of the time Sutton, Precup, & Singh, 1999 8 multi-step GTPs 4 stochastic primitive actions “Options” Precup 2000 Sutton, Precup, & Singh 1999 Predict: Probability of reaching each terminal hallway Goal: minimize # steps + values for target and other outcome hallway Policy Termination hallways Target (goal) hallway

27 Example: Open-the-door Predictor Use visual input to estimate  Probabilities of succeeding in opening the door, and of other outcomes (door locked, no handle, no real door)  expected cumulative cost (sub-par reward) in trying Experiment  Policy for walking up to the door, shaping grasp of handle, turning, pulling, and opening the door  Terminate on successful opening or various failure conditions  Measure outcome and cumulative cost

28 Example: RoboCup Soccer Pass Predictor uses perceived positions of ball, opponents, etc. to estimate probabilities of  Successful pass, openness of receiver  Interception  Reception failure  Aborted pass, in trouble  Aborted pass, something better to do  Loss of time Experiment  Policy for maneuvering ball, or around ball, to set up and pass  Termination strategy for aborting, recognizing completion  Measurement of outcome, time

29 Outline Experience as central to AI Predictive knowledge in General Generalized Transition Predictions (GTPs, or option models) Planning with GTPs (rooms-world example) State as predictions (PSRs) Prospects and conclusion

30 Combining Predictions If the mind is about predictions, Then thinking is combining predictions to produce new ones Predictions obviously compose  If A->B and B->C, then A->C GTPs are designed to do this generally  Fit into “Bellman equations” of semi-Markov extensions of dynamic programming  Can also be used for simulation-based planning

31 Composing Predictions A B B C A C Final measurement (e.g., partial distribution of outcome states) Transient measurement (e.g., elapsed time, cumulative reward)

32 Composing Predictions A B B C A C  1  1 then if B  2  2 T 1 .8T 2 B’.1 B’’.1.8 B’.1 B’’.1.8

33 Room-to-Room GTPs (General Transition Predictions) up down rightleft (to each room's 2 hallways) Fail 33% of the time Sutton, Precup, & Singh, 1999 8 multi-step GTPs 4 stochastic primitive actions “Options” Precup 2000 Sutton, Precup, & Singh 1999 Predict: Probability of reaching each terminal hallway Goal: minimize # steps + values for target and other outcome hallway Policy Termination hallways Target (goal) hallway

34 Planning with GTPs (GTPs)

35 Learning Path-to-Goal with and without GTPs Primitives GTPs & primitives

36 Rooms Example: Simultaneous Learning of all 8 GTPs from their Goals 0 0.1 0.2 0.3 0.4 020,00040,00060,00080,000100,000 All 8 hallway GTPs were learned accurately and efficiently while actions are selected totally at random goal prediction

37 Outline Experience as central to AI Predictive knowledge in General Generalized Transition Predictions (GTPs, or option models) Planning with GTPs (rooms-world example) State as predictions (PSRs) Prospects and conclusion

38 Predictive State Representations Problem: So far we have assumed states but world really just gives information, “observations” Hypothesis: What we normally think of as state is a set of predictions about outcomes of experiments  Wallet’s contents, John’s location, presence of objects… Prior work:  Learning deterministic FSAs - Rivest & Schapire, 1987  Adding stochasticity: An alternative to HMMs - Herbert Jaeger, 1999  Adding action: An alternative to POMDPs - Littman, Sutton, & Singh 2001

39 Summary of Results for Predictive State Rep’ns (PSRs) Exist compact, linear PSRs  # tests ≤ # states in minimal POMDP  # tests ≤ Rivest & Schapire’s Diversity  # tests can be exponentially fewer than diversity and POMDP Compact simulation/update process Construction algorithm from POMDP Learning/discovery algorithms of Rivest and Schapire, and of Jaeger, do not immediately extend to PSRs There are natural EM-like algorithms (current work)

40 Empty Gridworld with Local Sensing Four actions: Up, Down, Right, Left And four sensory bits

41 Distance to Wall Predictions 0 R 0 RR 1 RRR 1 RRRR... 0 D 1 DD 1 DDD... Predictive State Representation (PSR) 4 GTPs suffice to identify each state More needed to update PSR Many more are computed from PSR “meaning” of predictions

42 Suppose we add one non-uniformity 0 R 0 RR 1 RRR 1 RRRR... 0 D 1 DD 1 DDD... Now there is much more to know It would be challenging to program it all correctly

43 Other Extension Ideas Stochasticity Egocentric motion Multiple Rooms Second agent Moveable objects Transient goals It’s easy to make such problems arbitrarily challenging

44 Outline Experience as central to AI Predictive knowledge in general Generalized Transition Predictions (GTPs, or option models) Planning with GTPs (rooms-world example) State as predictions (PSRs) Prospects and conclusion

45 How Could These Ideas Proceed? Build systems! Build Gridworlds! A performance orientation would be problematic The “Knowledge Representation” guys may not be impressed But others I think will be very interested and appreciative - throughout modern probabalistic AI

46 The Experience Manifesto Experience is the input and output of AI An AI must have experience; it must have a life! Knowledge is about experience Not about objects, or people, or space, or time…except in so far as these things can be restated in terms of experience. Knowledge is well expressed as predictions of experience Predictions of experience have a much clearer meaning than any previously proposed kind of knowledge Predictions of experience can be autonomously verified Predictive knowledge is completely in the machine, not in a person! Planning is about composing predictions to search through the space of attainable experiences World-state rep’ns are also predictions of experience

47 Key Points We should not try to fake intelligence or understanding Computational Theory vs. just making it work  What to compute and why  Experience is central to AI  Knowledge should be about experience The minimal ontology Grounding in experience from the bottom up A computational theory of knowledge must support  Abstraction  Composition  Decomposition - Explicitness, verifiability Such Modularity is the whole point of knowledge

48 Summary of the Predictive View of AI Knowledge is Predictions About what-leads-to-what, under what ways of behaving Such knowledge is learnable, chainable Mental activity is working with predictions Learning them Combining them to produce new predictions (reasoning) Converting them to action (planning, reinforcement learning) Figuring out which are most useful Predictions are verifiable A natural way to self-maintain knowledge, which is essential for scaling AI beyond programming Most of the machinery is simple but potentially powerful Is it powerful enough?


Download ppt "Experience-Oriented Artificial Intelligence Rich Sutton with special thanks to Michael Littman, Doina Precup, Satinder Singh, David McAllester, Peter Stone,"

Similar presentations


Ads by Google