Handling non-determinism and incompleteness. Problems, Solutions, Success Measures: 3 orthogonal dimensions  Incompleteness in the initial state  Un.

Slides:



Advertisements
Similar presentations
Solving problems by searching
Advertisements

Heuristic Search techniques
REVIEW : Planning To make your thinking more concrete, use a real problem to ground your discussion. –Develop a plan for a person who is getting out of.
Problems and Their Classes
Planning with Non-Deterministic Uncertainty (Where failure is not an option) R&N: Chap. 12, Sect (+ Chap. 10, Sect 10.7)
Artificial Intelligence CS482, CS682, MW 1 – 2:15, SEM 201, MS 227 Prerequisites: 302, 365 Instructor: Sushil Louis,
Representing Boolean Functions for Symbolic Model Checking Supratik Chakraborty IIT Bombay.
CLASSICAL PLANNING What is planning ?  Planning is an AI approach to control  It is deliberation about actions  Key ideas  We have a model of the.
Planning based on Model Checking Dept. of Information Systems and Applied CS Bamberg University Seminar Paper Svetlana Balinova.
Automatic Verification Book: Chapter 6. What is verification? Traditionally, verification means proof of correctness automatic: model checking deductive:
Top 5 Worst Times For A Conference Talk 1.Last Day 2.Last Session of Last Day 3.Last Talk of Last Session of Last Day 4.Last Talk of Last Session of Last.
Friday: 2-3:15pm BY 510 make-up class Today: 1. Online search 2. Planning in Belief-space.
Situation Calculus for Action Descriptions We talked about STRIPS representations for actions. Another common representation is called the Situation Calculus.
Planning with Constraints Graphplan & SATplan Yongmei Shi.
Decision Theoretic Planning
Optimal Policies for POMDP Presented by Alp Sardağ.
4/22: Scheduling (contd) Planning with incomplete info (start) Earth which has many heights, and slopes and the unconfined plain that bind men together,
Planning CSE 473 Chapters 10.3 and 11. © D. Weld, D. Fox 2 Planning Given a logical description of the initial situation, a logical description of the.
Multi-agent Planning Amin Atrash. Papers Dynamic Planning for Multiple Mobile Robots –Barry L. Brummit, Anthony Stentz OBDD-based Universal Planning:
Planning under Uncertainty
Beyond Classical Search Non-Deterministic Actions  Transition model – Result(s,a) is no longer a singleton  Plans have to be “contingent”  Suck; if.
Partial Observability (State Uncertainty)  Assume non-determinism  Atomic model (for belief states and sensing actions)  Factored model (Progression/Regression)
Logic in general Logics are formal languages for representing information such that conclusions can be drawn Syntax defines the sentences in the language.
9/14: Belief Search Heuristics Today: Planning graph heuristics for belief search Wed: MDPs.
Planning Copyright, 1996 © Dale Carnegie & Associates, Inc. Chapter 11.
Problem Solving What is AI way of solving problem?
3/25  Monday 3/31 st 11:30AM BYENG 210 Talk by Dana Nau Planning for Interactions among Autonomous Agents.
3/27 Next big topic: Decision Theoretic Planning..
Nov 14 th  Homework 4 due  Project 4 due 11/26.
Handling non-determinism and incompleteness
5/6: Summary and Decision Theoretic Planning  Last homework socket opened (two more problems to be added—Scheduling, MDPs)  Project 3 due today  Sapa.
Markov Decision Processes
4/29: Conditional Planning  No Final. Instead we will have a last homework  Midterm to be returned Thursday; Homework reached Hanoi  Extra class on.
IT University of Copenhagen Lecture 8: Binary Decision Diagrams 1. Classical Boolean expression representations 2. If-then-else Normal Form (INF) 3. Binary.
Dynamic Bayesian Networks CSE 473. © Daniel S. Weld Topics Agency Problem Spaces Search Knowledge Representation Reinforcement Learning InferencePlanningLearning.
9/23. Announcements Homework 1 returned today (Avg 27.8; highest 37) –Homework 2 due Thursday Homework 3 socket to open today Project 1 due Tuesday –A.
Planning Where states are transparent and actions have preconditions and effects Notes at
Flavio Lerda 1 LTL Model Checking Flavio Lerda. 2 LTL Model Checking LTL –Subset of CTL* of the form: A f where f is a path formula LTL model checking.
Solving and Graphing Inequalities on a Number Line
Classical Planning Chapter 10.
15-820A 1 LTL to Büchi Automata Flavio Lerda A 2 LTL to Büchi Automata LTL Formulas Subset of CTL* –Distinct from CTL AFG p  LTL  f  CTL. f.
Digitaalsüsteemide verifitseerimise kursus1 Formal verification: BDD BDDs applied in equivalence checking.
Dana Nau: Lecture slides for Automated Planning Licensed under the Creative Commons Attribution-NonCommercial-ShareAlike License:
Dana Nau: Lecture slides for Automated Planning Licensed under the Creative Commons Attribution-NonCommercial-ShareAlike License:
Decision Making Under Uncertainty Lec #4: Planning and Sensing UIUC CS 598: Section EA Professor: Eyal Amir Spring Semester 2005 Uses slides by José Luis.
Planning with Non-Deterministic Uncertainty. Recap Uncertainty is inherent in systems that act in the real world Last lecture: reacting to unmodeled disturbances.
Simultaneously Learning and Filtering Juan F. Mancilla-Caceres CS498EA - Fall 2011 Some slides from Connecting Learning and Logic, Eyal Amir 2006.
Dana Nau: Lecture slides for Automated Planning Licensed under the Creative Commons Attribution-NonCommercial-ShareAlike License:
Lecture 3: 18/4/1435 Searching for solutions. Lecturer/ Kawther Abas 363CS – Artificial Intelligence.
Daniel Kroening and Ofer Strichman 1 Decision Procedures An Algorithmic Point of View BDDs.
AI Lecture 17 Planning Noémie Elhadad (substituting for Prof. McKeown)
Introduction to Artificial Intelligence Class 1 Planning & Search Henry Kautz Winter 2007.
Decision Theoretic Planning. Decisions Under Uncertainty  Some areas of AI (e.g., planning) focus on decision making in domains where the environment.
Intro to Planning Or, how to represent the planning problem in logic.
Dana Nau: Lecture slides for Automated Planning Licensed under the Creative Commons Attribution-NonCommercial-ShareAlike License:
Heuristic Search for problems with uncertainty CSE 574 April 22, 2003 Mausam.
Dana Nau: Lecture slides for Automated Planning Licensed under the Creative Commons Attribution-NonCommercial-ShareAlike License:
NPC.
Solving problems by searching A I C h a p t e r 3.
1 Solving problems by searching Chapter 3. 2 Outline Problem types Example problems Assumptions in Basic Search State Implementation Tree search Example.
The NP class. NP-completeness Lecture2. The NP-class The NP class is a class that contains all the problems that can be decided by a Non-Deterministic.
Solving problems by searching Chapter 3. Types of agents Reflex agent Consider how the world IS Choose action based on current percept Do not consider.
Solving problems by searching
Review for the Midterm Exam
Markov Decision Processes
Markov Decision Processes
Planning CSE 573 A handful of GENERAL SEARCH TECHNIQUES lie at the heart of practically all work in AI We will encounter the SAME PRINCIPLES again and.
Binary Decision Diagrams
Pendulum Swings in AI Top-down vs. Bottom-up
Reinforcement Learning Dealing with Partial Observability
Presentation transcript:

Handling non-determinism and incompleteness

Problems, Solutions, Success Measures: 3 orthogonal dimensions  Incompleteness in the initial state  Un (partial) observability of states  Non-deterministic actions  Uncertainty in state or effects  Complex reward functions (allowing degrees of satisfaction)  Conformant Plans: Don’t look— just do  Sequences  Contingent/Conditional Plans: Look, and based on what you see, Do; look again  Directed acyclic graphs  Policies: If in (belief) state S, do action a  (belief) state  action tables  Deterministic Success: Must reach goal-state with probability 1  Probabilistic Success: Must succeed with probability >= k (0<=k<=1)  Maximal Expected Reward: Maximize the expected reward (an optimization problem)

11/8

Some specific cases  1.0 success conformant planning for domains with incomplete initial states  1.0 success conformant planning for domains with non-deterministic actions  1.0 success conditional plans for fully observable domains with incompletely specified init states, and deterministic actions  1.0 success conditional plans for fully observable domains with non- deterministic actions  1.0 success conditional plans for parially observable domains with non- deterministic actions  Probabilistic variants of all the ones on the left (where we want success probability to be >= k).

Paths to Perdition Complexity of finding probability 1.0 success plans

Conformant Planning  Given an incomplete initial state, and a goal state, find a sequence of actions that when executed in any of the states consistent with the initial state, takes you to a goal state.  Belief State: is a set of states 2 S  I as well as G are belief states  (in classical planning, we already support partial goal state)  Issues:  Representation of Belief States  Generalizing “progression”, “regression” etc to belief states  Generating effective heuristics for estimating reachability in the space of belief states

Action Applicability Issue  Action applicability issue (what if a belief state has 100 states and an action is applicable to 90 of them?)  Consider actions that are always applicable in any state, but can leave many states unchanged.  This involves modeling actions without executability preconditions (they can have conditional effects). This ensures that the action is applicable everywhere

Generality of Belief State Rep

State Uncertainty and Actions  The size of a belief state B is the number of states in it.  For a world with k fluents, the size of a belief state can be between 1 (no uncertainty) and 2 k (complete uncertainty).  Actions applied to a belief state can both increase and reduce the size of a belief state  A non-deterministic action applied to a singleton belief state will lead to a larger (more uncertain) belief state  A deterministic action applied to a belief state can reduce its uncertainty  E.g. B={(pen-standing-on-table) (pen-on-ground)}; Action A is sweep the table. Effect is B’={(pen-on-ground)}  Often, a good heuristic in solving problems with large belief-state uncertainty is to do actions that reduce uncertainty  E.g. when you are blind-folded and left in the middle of a room, you try to reach the wall and then follow it to the door. Reaching the wall is a way of reducing your positional uncertainty

Progression and Regression with Belief States  Given a belief state B, and an action a, progression of B over a is defined as long as a is applicable in every state s in B  Progress(B,a)  { progress(s,a) | s in B}  Given a belief state B, and an action a, regression of B over a is defined as long as a is regressable from every state s in B.  Regress(B,a)  { regress(s,a) | s in B}  Non-deterministic actions complicate regression. Suppose an action a, when applied to state s can take us to s1 or s2 non-deterministically. Then, what is the regression of s1 over a ?  Strong and Weak pre-images: We consider B’ to be the strong pre-image of B w.r.t action a, if Progress(B’,a) is equal to B. We consider B’ to be a weak pre-image if Progress(B’, a ) is a superset of B

Belief State Search  Planning problem: initial belief state B I and goal state B G and a set of actions a i – the objective is to find a sequence of actions [a1…ak] that when executed in the initial belief state takes the agent to some state in B G  The plan is strong if every execution leads to a state in B G [probability of success is 1]  The plan is weak if some of the executions lead to a state in B G [probability of success > 0 ]  If we have stochastic actions, we can also talk about the “degree” of strength of the plan [ 0 <= p <= 1]  We will focus on STRONG plans  Search: Start with the initial belief state, B I and do progression or regression until you find a belief state B’ s.t. B’ is a subset of B G

Representing Belief States

Belief State Rep (cont)  Belief space planners have to search in the space of full propositional formulas!!  In contrast, classical state-space planners search in the space of interpretations (since states for classical planning were interpretations).  Several headaches:  Progression/Regression will have to be done over all states consistent with the formula (could be exponential number).  Checking for repeated search states will now involve checking the equivalence of logical formulas (aaugh..!)  To handle this problem, we have to convert the belief states into some canonical representation. We already know the CNF and DNF representations. There is another one, called Ordered Binary Decision Diagrams that is both canonical and compact OBDD can be thought of as a compact representation of the DNF version of the logical formula

Doing Progression/Regresssion Efficiently  Progression/Regression will have to be done over all states consistent with the formula (could be exponential number).  One way of handling this is to restrict the type of uncertainty allowed. For example, we may insist that every fluent must either be true, false or unknown. This will give us just the space of conjunctive logical formulas (only 3 n space).  Flip side is that we may not be able to represent all forms of uncertainty (e.g. how do we say that either P or Q is true in the initial state?)  Another idea is to directly manipulate the logical formulas during progression/regression (without expanding them into states…)  Tricky… connected to “Symbolic model checking”

Effective representations of logical formulas  Checking for repeated search states will now involve checking the equivalence of logical formulas (aaugh..!)  To handle this problem, we have to convert the belief states into some canonical representation.  We already know the CNF and DNF representations. These are normal forms but are not canonical  Same formula may have multiple equivalent CNF/DNF representations  There is another one, called Reduced Ordered Binary Decision Diagrams that is both canonical and compact ROBDD can be thought of as a compact representation of the DNF version of the logical formula

Symbolic model checking: The bird’s eye view  Belief states can be represented as logical formulas (and “implemented” as BDDs )  Transition functions can be represented as 2- stage logical formulas (and implemented as BDDs)  The operation of progressing a belief state through a transition function can be done entirely (and efficiently) in terms of operations on BDDs Read Appendix C before next class (emphasize C.5; C.6)

Conformant Planning: Efficiency Issues  Graphplan (CGP) and SAT-compilation approaches have also been tried for conformant planning  Idea is to make plan in one world, and try to extend it as needed to make it work in other worlds  Planning graph based heuristics for conformant planning have been investigated.  Interesting issues involving multiple planning graphs  Deriving Heuristics? – relaxed plans that work in multiple graphs  Compact representation? – Label graphs

KACMBP and Uncertainty reducing actions