Heuristic Search in MDPs 3/5/18
|
|
- Logan Richard
- 5 years ago
- Views:
Transcription
1 Heuristic Search in MDPs 3/5/18
2 Thinking about online planning. How can we use ideas we ve already seen to help with online planning? Heuristics? Iterative deepening? Monte Carlo simulations? Other ideas?
3 Heuristics What would happen if we started value iteration with non-zero initial values? Suppose we initialized values as follows:
4
5
6
7
8
9
10
11
12
13
14
15
16 LAO* Initialize graph with just the start state. A partial policy specifies actions for some states. If it s closed, it gives actions for all reachable states. Repeat until the optimal partial policy is closed: Expand a state that s reachable from a state that s reachable by the optimal partial policy. Update values that could have been affected by this expansion. Update the optimal partial policy.
17 Choosing Heuristics Suppose we had this MDP, where state 4 is terminal and has reward +1, while all other states are non-terminal and have reward 0. Actions succeed with probability ½ and fail (agent stays put) w/prob. ½. How could we initialize values to simplify the search? We want to make sure that LAO* doesn t bother fully exploring the path that starts by moving down.
18 <latexit sha1_base64="6l/mdi7gw32amgeuvyrwo8syzee=">aaab7nicbvdlsgnbeoynrxhfuy9ebomqeckuf/ugbl14jocaqlke2clsmmr2dp3pfulit3jxokjhv8ebf+pkcddegoaiqpvurjcvwqdrfju5pewv1bx8emfjc2t7p7i7d2+stdpus0qmuhfsw6vq3eebkjdszwkcsl4p+9djv/7itrgjusnbyooydpwibknopuavbi7jjxhbxzjbcscgi8sbkvl1shxyaqc1dvgr1ulyfnoftfjjmp6byjckggwtffrozyanlpvplzctvttmjhho7h2ri6t0sjrowwrjrp09masxmym4tj0xxz6z98bif14zw+g8gaqvzsgvmy6kmkkwiepnsudozlaolkfmc3sryt2qkumbucgg4m2/vej808pfxbu1yvzbfhk4gemogwdnuiubqiepdcq8wqu8og/os/pmve9bc85szh/+wpn8ayqpj/w=</latexit> <latexit sha1_base64="venw3qw2t68dlkdtvu7pt0wj9go=">aaab7nicbvbns8naej3ur1q/qh69lc1crsijf/ugfl14rgbsoq1ls920szebulsrquifemgdild/j7f+gzdtd9r6yodx3gwz8/yym6vte2ivvlbx1jekm6wt7z3dvfl+wyokekmosyieybapfevmufczzwk7lhshpqctf3st+60nkhwlxl1oy+qfecbywajwrmopa+oexsg7v67adxskteycoak2kt3tl0kjbfbk391+rjkqck04vqrj2lh2miw1i5yos91e0riter7qjqech1r52ftemto2sh8fktqlnjqqvycyhcqvhr7pdleeqkuvf//zookolrymitjrvjdzoidhsecofx71mare89qqtcqztyiyxbitbsiqmrccxzexixtwv6w7dyama5ihcedqgro4ca4nuiumuecawzo8wbv1al1ah9bnrlvgzwco4q+srx+mdzgc</latexit> <latexit sha1_base64="venw3qw2t68dlkdtvu7pt0wj9go=">aaab7nicbvbns8naej3ur1q/qh69lc1crsijf/ugfl14rgbsoq1ls920szebulsrquifemgdild/j7f+gzdtd9r6yodx3gwz8/yym6vte2ivvlbx1jekm6wt7z3dvfl+wyokekmosyieybapfevmufczzwk7lhshpqctf3st+60nkhwlxl1oy+qfecbywajwrmopa+oexsg7v67adxskteycoak2kt3tl0kjbfbk391+rjkqck04vqrj2lh2miw1i5yos91e0riter7qjqech1r52ftemto2sh8fktqlnjqqvycyhcqvhr7pdleeqkuvf//zookolrymitjrvjdzoidhsecofx71mare89qqtcqztyiyxbitbsiqmrccxzexixtwv6w7dyama5ihcedqgro4ca4nuiumuecawzo8wbv1al1ah9bnrlvgzwco4q+srx+mdzgc</latexit> <latexit sha1_base64="2cwvhlz2mxneqyw662c40ph3eog=">aaab7nicbva9swnbej2lxzf+rs1tfomqm3bnoxzc0myygmccsqh7m71kyd7eutsnhcn/wszcxdbfy+e/czncoykpbh7vztazl0ikmoi6305hzxvtfao4wdra3tndk+8fpjg41yz7ljaxbgxuccku91gg5k1ecxofkjed0c3ubz5xbuss7ngc8g5eb0qeglg0umtynafkiri9cswtutoqzellpai5gr3yv6cfszticpmkxrq9n8furjukjvmk1ekntygb0qfvw6poxe03m907isdw6zmw1ryukpn6eykjkthjklcdecwhwfsm4n9eo8xwopsjlatifzsvclnjmcbt50lfam5qji2htat7k2fdqildg1hjhuatvrxm/lpazc27cyv16zynihzbmvtbg3oowy00wacgep7hfd6cr+ffexc+5q0fj585hd9wpn8aczoocw==</latexit> <latexit sha1_base64="bcidou0jxf+dw6x0mmcelvo6fsi=">aaab93icbzdntsjafivv8q/xh6plnxojcbogrtfrd0q3ljgxqakvticptjhom5mpcty8irsxatz6ku58gwfoqsgttpll3htz75wg4uxpx/m2ciura+sbxc3s1vbobtne22+qojweeitmswwhwfhobpu005y2e0lxfhdacky303rrkurfyngvxwn1izwqlgqea2p17pkwqk5qd0br8+huym+uodvnjrqmbg4vynxo2v/dfkzsiapnofaq4zqj9jmsnsoctkrdvneekxee0i5bgsoq/gx2+aqdg6epwliajzsaub8nmhwpny4c0xlhpvsltan5x62t6vdsz5hiuk0fms8ku450jkypod6tlgg+nocjzozwrizyyqjnviutgrv45wxwzmpxnffuvfk/ztmowiecqrvcuia63eidpccqwjo8wpv1zl1y79bhvlvg5tmh8efw5w+yhjfl</latexit> <latexit sha1_base64="bcidou0jxf+dw6x0mmcelvo6fsi=">aaab93icbzdntsjafivv8q/xh6plnxojcbogrtfrd0q3ljgxqakvticptjhom5mpcty8irsxatz6ku58gwfoqsgttpll3htz75wg4uxpx/m2ciura+sbxc3s1vbobtne22+qojweeitmswwhwfhobpu005y2e0lxfhdacky303rrkurfyngvxwn1izwqlgqea2p17pkwqk5qd0br8+huym+uodvnjrqmbg4vynxo2v/dfkzsiapnofaq4zqj9jmsnsoctkrdvneekxee0i5bgsoq/gx2+aqdg6epwliajzsaub8nmhwpny4c0xlhpvsltan5x62t6vdsz5hiuk0fms8ku450jkypod6tlgg+nocjzozwrizyyqjnviutgrv45wxwzmpxnffuvfk/ztmowiecqrvcuia63eidpccqwjo8wpv1zl1y79bhvlvg5tmh8efw5w+yhjfl</latexit> <latexit sha1_base64="bcidou0jxf+dw6x0mmcelvo6fsi=">aaab93icbzdntsjafivv8q/xh6plnxojcbogrtfrd0q3ljgxqakvticptjhom5mpcty8irsxatz6ku58gwfoqsgttpll3htz75wg4uxpx/m2ciura+sbxc3s1vbobtne22+qojweeitmswwhwfhobpu005y2e0lxfhdacky303rrkurfyngvxwn1izwqlgqea2p17pkwqk5qd0br8+huym+uodvnjrqmbg4vynxo2v/dfkzsiapnofaq4zqj9jmsnsoctkrdvneekxee0i5bgsoq/gx2+aqdg6epwliajzsaub8nmhwpny4c0xlhpvsltan5x62t6vdsz5hiuk0fms8ku450jkypod6tlgg+nocjzozwrizyyqjnviutgrv45wxwzmpxnffuvfk/ztmowiecqrvcuia63eidpccqwjo8wpv1zl1y79bhvlvg5tmh8efw5w+yhjfl</latexit> <latexit sha1_base64="bcidou0jxf+dw6x0mmcelvo6fsi=">aaab93icbzdntsjafivv8q/xh6plnxojcbogrtfrd0q3ljgxqakvticptjhom5mpcty8irsxatz6ku58gwfoqsgttpll3htz75wg4uxpx/m2ciura+sbxc3s1vbobtne22+qojweeitmswwhwfhobpu005y2e0lxfhdacky303rrkurfyngvxwn1izwqlgqea2p17pkwqk5qd0br8+huym+uodvnjrqmbg4vynxo2v/dfkzsiapnofaq4zqj9jmsnsoctkrdvneekxee0i5bgsoq/gx2+aqdg6epwliajzsaub8nmhwpny4c0xlhpvsltan5x62t6vdsz5hiuk0fms8ku450jkypod6tlgg+nocjzozwrizyyqjnviutgrv45wxwzmpxnffuvfk/ztmowiecqrvcuia63eidpccqwjo8wpv1zl1y79bhvlvg5tmh8efw5w+yhjfl</latexit> Admissible Heuristics What constitutes an admissible heuristic for LAO*? In A* search, admissibility guarantees that the optimal path will be found. For LAO*, we want to ensure that an optimal closed partial policy is found. If s is terminal: h(s) =0 A partial policy specifies actions for some states. If it s closed, it gives actions for all reachable states. Otherwise: h(s) V (s)
19 Real-Time Dynamic Programming Repeat while there s time remaining: state ß start state What does admissibility guarantee in RTDP? repeat until terminal (or depth bound): action ß optimal action in current state V(state) ß R(state) + discount * Q(state, action) Q(state, action) calculated from V(s ) for all reachable s. If s hasn t been seen before, initialize V(s ) ß h(s ). state ß result of taking action
20 Online Planning An online planner is one that interleaves planning and acting. Are LAO* and RTDP online planners? If not, how could we modify them to work online?
Monte Carlo Tree Search
Monte Carlo Tree Search Branislav Bošanský PAH/PUI 2016/2017 MDPs Using Monte Carlo Methods Monte Carlo Simulation: a technique that can be used to solve a mathematical or statistical problem using repeated
More informationClass Overview. Introduction to Artificial Intelligence COMP 3501 / COMP Lecture 2: Search. Problem Solving Agents
Class Overview COMP 3501 / COMP 4704-4 Lecture 2: Search Prof. 1 2 Problem Solving Agents Problem Solving Agents: Assumptions Requires a goal Assume world is: Requires actions Observable What actions?
More informationName: UW CSE 473 Midterm, Fall 2014
Instructions Please answer clearly and succinctly. If an explanation is requested, think carefully before writing. Points may be removed for rambling answers. If a question is unclear or ambiguous, feel
More informationLearning Objectives. c D. Poole and A. Mackworth 2010 Artificial Intelligence, Lecture 3.5, Page 1
Learning Objectives At the end of the class you should be able to: justify why depth-bounded search is useful demonstrate how iterative-deepening works for a particular problem demonstrate how depth-first
More informationProblem Solving and Search
Artificial Intelligence Problem Solving and Search Dae-Won Kim School of Computer Science & Engineering Chung-Ang University Outline Problem-solving agents Problem types Problem formulation Example problems
More informationInformed (Heuristic) Search. Idea: be smart about what paths to try.
Informed (Heuristic) Search Idea: be smart about what paths to try. 1 Blind Search vs. Informed Search What s the difference? How do we formally specify this? A node is selected for expansion based on
More informationMonte Carlo Tree Search PAH 2015
Monte Carlo Tree Search PAH 2015 MCTS animation and RAVE slides by Michèle Sebag and Romaric Gaudel Markov Decision Processes (MDPs) main formal model Π = S, A, D, T, R states finite set of states of the
More informationApproximate Q-Learning 3/23/18
Approximate Q-Learning 3/23/18 On-Policy Learning (SARSA) Instead of updating based on the best action from the next state, update based on the action your current policy actually takes from the next state.
More informationClass Overview. Introduction to Artificial Intelligence COMP 3501 / COMP Lecture 2. Problem Solving Agents. Problem Solving Agents: Assumptions
Class Overview COMP 3501 / COMP 4704-4 Lecture 2 Prof. JGH 318 Problem Solving Agents Problem Solving Agents: Assumptions Requires a goal Assume world is: Requires actions Observable What actions? Discrete
More informationSearching with Partial Information
Searching with Partial Information Above we (unrealistically) assumed that the environment is fully observable and deterministic Moreover, we assumed that the agent knows what the effects of each action
More information1 Introduction and Examples
1 Introduction and Examples Sequencing Problems Definition A sequencing problem is one that involves finding a sequence of steps that transforms an initial system state to a pre-defined goal state for
More informationAdvanced A* Improvements
Advanced A* Improvements 1 Iterative Deepening A* (IDA*) Idea: Reduce memory requirement of A* by applying cutoff on values of f Consistent heuristic function h Algorithm IDA*: 1. Initialize cutoff to
More informationMonte Carlo Tree Search
Monte Carlo Tree Search 2-15-16 Reading Quiz What is the relationship between Monte Carlo tree search and upper confidence bound applied to trees? a) MCTS is a type of UCB b) UCB is a type of MCTS c) both
More informationThe exam is closed book, closed calculator, and closed notes except your one-page crib sheet.
CS Summer Introduction to Artificial Intelligence Midterm You have approximately minutes. The exam is closed book, closed calculator, and closed notes except your one-page crib sheet. Mark your answers
More informationInformed Search Algorithms
Informed Search Algorithms CITS3001 Algorithms, Agents and Artificial Intelligence Tim French School of Computer Science and Software Engineering The University of Western Australia 2017, Semester 2 Introduction
More informationA.I.: Informed Search Algorithms. Chapter III: Part Deux
A.I.: Informed Search Algorithms Chapter III: Part Deux Best-first search Greedy best-first search A * search Heuristics Outline Overview Informed Search: uses problem-specific knowledge. General approach:
More informationCSC384 Midterm Sample Questions
CSC384 Midterm Sample Questions Fall 2016 1 Search Short Answer 1. It would seem that iterative deepening search should have a higher asymptotic time complexity than breadth-first search because every
More informationSearch : Lecture 2. September 9, 2003
Search 6.825: Lecture 2 September 9, 2003 1 Problem-Solving Problems When your environment can be effectively modeled as having discrete states and actions deterministic, known world dynamics known initial
More informationProbabilistic Planning with Markov Decision Processes. Andrey Kolobov and Mausam Computer Science and Engineering University of Washington, Seattle
Probabilistic Planning with Markov Decision Processes Andrey Kolobov and Mausam Computer Science and Engineering University of Washington, Seattle 1 Goal an extensive introduction to theory and algorithms
More informationArtificial Intelligence
Artificial Intelligence Search Marc Toussaint University of Stuttgart Winter 2015/16 (slides based on Stuart Russell s AI course) Outline Problem formulation & examples Basic search algorithms 2/100 Example:
More informationTo earn the extra credit, one of the following has to hold true. Please circle and sign.
CS 188 Spring 2011 Introduction to Artificial Intelligence Practice Final Exam To earn the extra credit, one of the following has to hold true. Please circle and sign. A I spent 3 or more hours on the
More informationCS 540-1: Introduction to Artificial Intelligence
CS 540-1: Introduction to Artificial Intelligence Exam 1: 7:15-9:15pm, October 11, 1995 CLOSED BOOK (one page of notes allowed) Write your answers on these pages and show your work. If you feel that a
More informationProblem Spaces & Search CSE 473
Problem Spaces & Search Problem Spaces & Search CSE 473 473 Topics 473 Topics Agents & Environments Problem Spaces Search & Constraint Satisfaction Knowledge Repr n & Logical Reasoning Machine Learning
More informationInformed State Space Search B4B36ZUI, LS 2018
Informed State Space Search B4B36ZUI, LS 2018 Branislav Bošanský, Martin Schaefer, David Fiedler, Jaromír Janisch {name.surname}@agents.fel.cvut.cz Artificial Intelligence Center, Czech Technical University
More informationPlanning for Markov Decision Processes with Sparse Stochasticity
Planning for Markov Decision Processes with Sparse Stochasticity Maxim Likhachev Geoff Gordon Sebastian Thrun School of Computer Science School of Computer Science Dept. of Computer Science Carnegie Mellon
More informationHeuristic Search. Robert Platt Northeastern University. Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA
Heuristic Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA Recap: What is graph search? Start state Goal state Graph search: find a path
More informationToday s s lecture. Lecture 3: Search - 2. Problem Solving by Search. Agent vs. Conventional AI View. Victor R. Lesser. CMPSCI 683 Fall 2004
Today s s lecture Search and Agents Material at the end of last lecture Lecture 3: Search - 2 Victor R. Lesser CMPSCI 683 Fall 2004 Continuation of Simple Search The use of background knowledge to accelerate
More informationHeuristic Search in Cyclic AND/OR Graphs
From: AAAI-98 Proceedings. Copyright 1998, AAAI (www.aaai.org). All rights reserved. Heuristic Search in Cyclic AND/OR Graphs Eric A. Hansen and Shlomo Zilberstein Computer Science Department University
More informationPlanning and Control: Markov Decision Processes
CSE-571 AI-based Mobile Robotics Planning and Control: Markov Decision Processes Planning Static vs. Dynamic Predictable vs. Unpredictable Fully vs. Partially Observable Perfect vs. Noisy Environment What
More informationSymbolic LAO* Search for Factored Markov Decision Processes
Symbolic LAO* Search for Factored Markov Decision Processes Zhengzhu Feng Computer Science Department University of Massachusetts Amherst MA 01003 Eric A. Hansen Computer Science Department Mississippi
More informationUninformed Search Strategies AIMA
Uninformed Search Strategies AIMA 3.3-3.4 CIS 421/521 - Intro to AI - Fall 2017 1 Review: Formulating search problems Formulate search problem States: configurations of the puzzle (9! configurations) Actions:
More informationReinforcement Learning: A brief introduction. Mihaela van der Schaar
Reinforcement Learning: A brief introduction Mihaela van der Schaar Outline Optimal Decisions & Optimal Forecasts Markov Decision Processes (MDPs) States, actions, rewards and value functions Dynamic Programming
More informationˆ The exam is closed book, closed calculator, and closed notes except your one-page crib sheet.
CS Summer Introduction to Artificial Intelligence Midterm ˆ You have approximately minutes. ˆ The exam is closed book, closed calculator, and closed notes except your one-page crib sheet. ˆ Mark your answers
More informationHeuristic (Informed) Search
Heuristic (Informed) Search (Where we try to choose smartly) R&N: Chap., Sect..1 3 1 Search Algorithm #2 SEARCH#2 1. INSERT(initial-node,Open-List) 2. Repeat: a. If empty(open-list) then return failure
More informationCIS 192: Artificial Intelligence. Search and Constraint Satisfaction Alex Frias Nov. 30 th
CIS 192: Artificial Intelligence Search and Constraint Satisfaction Alex Frias Nov. 30 th What is AI? Designing computer programs to complete tasks that are thought to require intelligence 4 categories
More informationHeuristic Search. Rob Platt Northeastern University. Some images and slides are used from: AIMA
Heuristic Search Rob Platt Northeastern University Some images and slides are used from: AIMA Recap: What is graph search? Start state Goal state Graph search: find a path from start to goal what are the
More informationDecision making for autonomous naviga2on. Anoop Aroor Advisor: Susan Epstein CUNY Graduate Center, Computer science
Decision making for autonomous naviga2on Anoop Aroor Advisor: Susan Epstein CUNY Graduate Center, Computer science Overview Naviga2on and Mobile robots Decision- making techniques for naviga2on Building
More informationBest-First Search Minimizing Space or Time. IDA* Save space, take more time
Best-First Search Minimizing Space or Time IDA* Save space, take more time IDA*-1 A* space complexity» What does the space complexity of A* depend upon? IDA*-2 A* space complexity 2» What does the space
More informationSolving problems by searching
Solving problems by searching Chapter 3 Some slide credits to Hwee Tou Ng (Singapore) Outline Problem-solving agents Problem types Problem formulation Example problems Basic search algorithms Heuristics
More informationTwo-player Games ZUI 2016/2017
Two-player Games ZUI 2016/2017 Branislav Bošanský bosansky@fel.cvut.cz Two Player Games Important test environment for AI algorithms Benchmark of AI Chinook (1994/96) world champion in checkers Deep Blue
More informationAssignment 1 is out! Due: 9 Sep 23:59! Can work in a group of 2-3 students.! NO cheating!!!! Submit in turnitin! Code + report!
Assignment 1 is out! Due: 9 Sep 23:59! Submit in turnitin! Code + report! Can work in a group of 2-3 students.! Please register your group in the website linked from the assignment description before tomorrow
More informationLecture 5 Heuristics. Last Time: A* Search
CSE 473 Lecture 5 Heuristics CSE AI Faculty Last Time: A* Search Use an evaluation function f(n) for node n. f(n) = estimated total cost of path thru n to goal f(n) = g(n) + h(n) g(n) = cost so far to
More informationSiphon: Expediting Inter-Datacenter Coflows in Wide-Area Data Analytics. Shuhao Liu, Li Chen, Baochun Li University of Toronto July 12, 2018
Siphon: Expediting Inter-Datacenter Coflows in Wide-Area Data Analytics Shuhao Liu, Li Chen, Baochun Li University of Toronto July 12, 2018 What is a Coflow? One stage in a data analytic job Map 1 Reduce
More informationProblem Solving & Heuristic Search
190.08 Artificial 2016-Spring Problem Solving & Heuristic Search Byoung-Tak Zhang School of Computer Science and Engineering Seoul National University 190.08 Artificial (2016-Spring) http://www.cs.duke.edu/courses/fall08/cps270/
More informationEfficient memory-bounded search methods
Efficient memory-bounded search methods Mikhail Simin Arjang Fahim CSCE 580: Artificial Intelligence Fall 2011 Dr. Marco Voltorta Outline of The Presentation Motivations and Objectives Background - BFS
More informationLAO*, RLAO*, or BLAO*?
, R, or B? Peng Dai and Judy Goldsmith Computer Science Dept. University of Kentucky 773 Anderson Tower Lexington, KY 40506-0046 Abstract In 2003, Bhuma and Goldsmith introduced a bidirectional variant
More informationUninformed Search. Reading: Chapter 4 (Tuesday, 2/5) HW#1 due next Tuesday
Uninformed Search Reading: Chapter 4 (Tuesday, 2/5) HW#1 due next Tuesday 1 Uninformed Search through the space of possible solutions Use no knowledge about which path is likely to be best Exception: uniform
More informationARTIFICIAL INTELLIGENCE
3010 ARTIFICIAL INTELLIGENCE Lecture 4 Iterative Deepening A* Masashi Shimbo 2017-05-02 Depth-first search Recursive implementation Depth-first search can be implemented in a recursive fashion 2 / 44 Depth-first
More informationUsing Domain-Configurable Search Control for Probabilistic Planning
Using Domain-Configurable Search Control for Probabilistic Planning Ugur Kuter and Dana Nau Department of Computer Science and Institute for Systems Research University of Maryland College Park, MD 20742,
More informationLast time: Problem-Solving
Last time: Problem-Solving Problem solving: Goal formulation Problem formulation (states, operators) Search for solution Problem formulation: Initial state??? 1 Last time: Problem-Solving Problem types:
More informationAn Overview of MDP Planning Research at ICAPS. Andrey Kolobov and Mausam Computer Science and Engineering University of Washington, Seattle
An Overview of MDP Planning Research at ICAPS Andrey Kolobov and Mausam Computer Science and Engineering University of Washington, Seattle 1 What is ICAPS? International Conference on Automated Planning
More informationHeuristic Search. CPSC 470/570 Artificial Intelligence Brian Scassellati
Heuristic Search CPSC 470/570 Artificial Intelligence Brian Scassellati Goal Formulation 200 Denver 300 200 200 Chicago 150 200 Boston 50 1200 210 75 320 255 Key West New York Well-defined function that
More informationGradient Descent. 1) S! initial state 2) Repeat: Similar to: - hill climbing with h - gradient descent over continuous space
Local Search 1 Local Search Light-memory search method No search tree; only the current state is represented! Only applicable to problems where the path is irrelevant (e.g., 8-queen), unless the path is
More informationUsing Domain-Configurable Search Control for Probabilistic Planning
Using Domain-Configurable Search Control for Probabilistic Planning Ugur Kuter and Dana Nau Department of Computer Science and Institute for Systems Research University of Maryland College Park, MD 20742,
More informationFinal Exam. Introduction to Artificial Intelligence. CS 188 Spring 2010 INSTRUCTIONS. You have 3 hours.
CS 188 Spring 2010 Introduction to Artificial Intelligence Final Exam INSTRUCTIONS You have 3 hours. The exam is closed book, closed notes except a two-page crib sheet. Please use non-programmable calculators
More informationCSE 473. Chapter 4 Informed Search. CSE AI Faculty. Last Time. Blind Search BFS UC-BFS DFS DLS Iterative Deepening Bidirectional Search
CSE 473 Chapter 4 Informed Search CSE AI Faculty Blind Search BFS UC-BFS DFS DLS Iterative Deepening Bidirectional Search Last Time 2 1 Repeated States Failure to detect repeated states can turn a linear
More informationCS 331: Artificial Intelligence Informed Search. Informed Search
CS 331: Artificial Intelligence Informed Search 1 Informed Search How can we make search smarter? Use problem-specific knowledge beyond the definition of the problem itself Specifically, incorporate knowledge
More informationCS 495 & 540: Problem Set 1
CS 495 & 540: Problem Set 1 Section: MW 10-11:50 am Total: 150pts Due: 02/10/2016 Instructions: 1. I leave plenty of space on each page for your computation. If you need more sheet, please attach your
More informationMidterm Examination CS 540-2: Introduction to Artificial Intelligence
Midterm Examination CS 54-2: Introduction to Artificial Intelligence March 9, 217 LAST NAME: FIRST NAME: Problem Score Max Score 1 15 2 17 3 12 4 6 5 12 6 14 7 15 8 9 Total 1 1 of 1 Question 1. [15] State
More informationSearch: Advanced Topics and Conclusion
Search: Advanced Topics and Conclusion CPSC 322 Lecture 8 January 20, 2006 Textbook 2.6 Search: Advanced Topics and Conclusion CPSC 322 Lecture 8, Slide 1 Lecture Overview Recap Branch & Bound A Tricks
More informationCS 331: Artificial Intelligence Informed Search. Informed Search
CS 331: Artificial Intelligence Informed Search 1 Informed Search How can we make search smarter? Use problem-specific knowledge beyond the definition of the problem itself Specifically, incorporate knowledge
More informationKünstliche Intelligenz
Künstliche Intelligenz 3. Suche Teil 3 Dr. Claudia Schon schon@uni-koblenz.de Arbeitsgruppe Künstliche Intelligenz Universität Koblenz-Landau Except for some small changes these slides are transparencies
More informationReinforcement Learning (INF11010) Lecture 6: Dynamic Programming for Reinforcement Learning (extended)
Reinforcement Learning (INF11010) Lecture 6: Dynamic Programming for Reinforcement Learning (extended) Pavlos Andreadis, February 2 nd 2018 1 Markov Decision Processes A finite Markov Decision Process
More informationUninformed Search Methods
Uninformed Search Methods Search Algorithms Uninformed Blind search Breadth-first uniform first depth-first Iterative deepening depth-first Bidirectional Branch and Bound Informed Heuristic search Greedy
More informationThis lecture. Lecture 6: Search 5. Other Time and Space Variations of A* Victor R. Lesser. RBFS - Recursive Best-First Search Algorithm
Lecture 6: Search 5 Victor R. Lesser CMPSCI 683 Fall 2010 This lecture Other Time and Space Variations of A* Finish off RBFS SMA* Anytime A* RTA* (maybe if have time) RBFS - Recursive Best-First Search
More informationPotential Midterm Exam Questions
Potential Midterm Exam Questions 1. What are the four ways in which AI is usually viewed? Which of the four is the preferred view of the authors of our textbook? 2. What does each of the lettered items
More informationInformed Search. Xiaojin Zhu Computer Sciences Department University of Wisconsin, Madison
Informed Search Xiaojin Zhu jerryzhu@cs.wisc.edu Computer Sciences Department University of Wisconsin, Madison [Based on slides from Andrew Moore http://www.cs.cmu.edu/~awm/tutorials ] slide 1 Main messages
More informationIntelligent Agents. Planning Graphs - The Graph Plan Algorithm. Ute Schmid. Cognitive Systems, Applied Computer Science, Bamberg University
Intelligent Agents Planning Graphs - The Graph Plan Algorithm Ute Schmid Cognitive Systems, Applied Computer Science, Bamberg University last change: July 9, 2015 U. Schmid (CogSys) Intelligent Agents
More informationCMU-Q Lecture 2: Search problems Uninformed search. Teacher: Gianni A. Di Caro
CMU-Q 15-381 Lecture 2: Search problems Uninformed search Teacher: Gianni A. Di Caro RECAP: ACT RATIONALLY Think like people Think rationally Agent Sensors? Actuators Percepts Actions Environment Act like
More informationArtificial Intelligence
Artificial Intelligence Informed Search and Exploration Chapter 4 (4.1 4.2) A General Search algorithm: Chapter 3: Search Strategies Task : Find a sequence of actions leading from the initial state to
More informationCS 416, Artificial Intelligence Midterm Examination Fall 2004
CS 416, Artificial Intelligence Midterm Examination Fall 2004 Name: This is a closed book, closed note exam. All questions and subquestions are equally weighted. Introductory Material 1) True or False:
More informationInformed search strategies (Section ) Source: Fotolia
Informed search strategies (Section 3.5-3.6) Source: Fotolia Review: Tree search Initialize the frontier using the starting state While the frontier is not empty Choose a frontier node to expand according
More informationMidterm I. Introduction to Artificial Intelligence. CS 188 Fall You have approximately 3 hours.
CS 88 Fall 202 Introduction to Artificial Intelligence Midterm I You have approximately 3 hours. The exam is closed book, closed notes except a one-page crib sheet. Please use non-programmable calculators
More informationIntroduction to Fall 2008 Artificial Intelligence Midterm Exam
CS 188 Introduction to Fall 2008 Artificial Intelligence Midterm Exam INSTRUCTIONS You have 80 minutes. 70 points total. Don t panic! The exam is closed book, closed notes except a one-page crib sheet,
More informationMonotonicity. Admissible Search: That finds the shortest path to the Goal. Monotonicity: local admissibility is called MONOTONICITY
Monotonicity Admissible Search: That finds the shortest path to the Goal Monotonicity: local admissibility is called MONOTONICITY This property ensures consistently minimal path to each state they encounter
More informationMidterm I. Introduction to Artificial Intelligence. CS 188 Fall You have approximately 3 hours.
CS 88 Fall 202 Introduction to Artificial Intelligence Midterm I You have approximately 3 hours. The exam is closed book, closed notes except a one-page crib sheet. Please use non-programmable calculators
More informationAlgorithms for Solving RL: Temporal Difference Learning (TD) Reinforcement Learning Lecture 10
Algorithms for Solving RL: Temporal Difference Learning (TD) 1 Reinforcement Learning Lecture 10 Gillian Hayes 8th February 2007 Incremental Monte Carlo Algorithm TD Prediction TD vs MC vs DP TD for control:
More informationARTIFICIAL INTELLIGENCE (CSC9YE ) LECTURES 2 AND 3: PROBLEM SOLVING
ARTIFICIAL INTELLIGENCE (CSC9YE ) LECTURES 2 AND 3: PROBLEM SOLVING BY SEARCH Gabriela Ochoa http://www.cs.stir.ac.uk/~goc/ OUTLINE Problem solving by searching Problem formulation Example problems Search
More informationSolving Problem by Searching. Chapter 3
Solving Problem by Searching Chapter 3 Outline Problem-solving agents Problem formulation Example problems Basic search algorithms blind search Heuristic search strategies Heuristic functions Problem-solving
More information: Principles of Automated Reasoning and Decision Making Midterm
16.410-13: Principles of Automated Reasoning and Decision Making Midterm October 20 th, 2003 Name E-mail Note: Budget your time wisely. Some parts of this quiz could take you much longer than others. Move
More informationOn-line search for solving Markov Decision Processes via heuristic sampling
On-line search for solving Markov Decision Processes via heuristic sampling Laurent Péret and Frédérick Garcia Abstract. In the past, Markov Decision Processes (MDPs) have become a standard for solving
More informationPartially Observable Markov Decision Processes. Silvia Cruciani João Carvalho
Partially Observable Markov Decision Processes Silvia Cruciani João Carvalho MDP A reminder: is a set of states is a set of actions is the state transition function. is the probability of ending in state
More informationAI: Week 2. Tom Henderson. Fall 2014 CS 5300
AI: Week 2 Tom Henderson Fall 2014 What s a Problem? Initial state Actions Transition model Goal Test Path Cost Does this apply to: Problem: Get A in CS5300 Solution: action sequence from initial to goal
More information3 SOLVING PROBLEMS BY SEARCHING
48 3 SOLVING PROBLEMS BY SEARCHING A goal-based agent aims at solving problems by performing actions that lead to desirable states Let us first consider the uninformed situation in which the agent is not
More informationClick to edit Master title style Approximate Models for Batch RL Click to edit Master subtitle style Emma Brunskill 2/18/15 2/18/15 1 1
Approximate Click to edit Master titlemodels style for Batch RL Click to edit Emma Master Brunskill subtitle style 11 FVI / FQI PI Approximate model planners Policy Iteration maintains both an explicit
More informationsearch, DFS & BrFS; cycle checking & MPC arc costs; heuristics; LCFS, BeFS, A* misc: iterative deepening, etc.
CSC384: Lecture 5 Last time search, DFS & BrFS; cycle checking & MPC Today arc costs; heuristics; LCFS, BeFS, A* misc: iterative deepening, etc. Readings: Today: Ch.4.5, 4.6 Next Weds: class notes (no
More informationCSC384 Test 1 Sample Questions
CSC384 Test 1 Sample Questions October 27, 2015 1 Short Answer 1. Is A s search behavior necessarily exponentially explosive?. That is, does its search time always grow at least exponentially with the
More informationInformed Search. CS 486/686 University of Waterloo May 10. cs486/686 Lecture Slides 2005 (c) K. Larson and P. Poupart
Informed Search CS 486/686 University of Waterloo May 0 Outline Using knowledge Heuristics Best-first search Greedy best-first search A* search Other variations of A* Back to heuristics 2 Recall from last
More informationOutline for today s lecture. Informed Search I. One issue: How to search backwards? Very briefly: Bidirectional search. Outline for today s lecture
Outline for today s lecture Informed Search I Uninformed Search Briefly: Bidirectional Search (AIMA 3.4.6) Uniform Cost Search (UCS) Informed Search Introduction to Informed search Heuristics 1 st attempt:
More informationCS 4700: Artificial Intelligence
CS 4700: Foundations of Artificial Intelligence Fall 2017 Instructor: Prof. Haym Hirsh Lecture 7 Extra Credit Opportunity: Lecture Today 4:15pm Gates G01 Learning to See Without a Teacher Phillip Isola
More informationInformed Search Methods
Informed Search Methods How can we improve searching strategy by using intelligence? Map example: Heuristic: Expand those nodes closest in as the crow flies distance to goal 8-puzzle: Heuristic: Expand
More informationFast optimal task graph scheduling by means of an optimized parallel A -Algorithm
Fast optimal task graph scheduling by means of an optimized parallel A -Algorithm Udo Hönig and Wolfram Schiffmann FernUniversität Hagen, Lehrgebiet Rechnerarchitektur, 58084 Hagen, Germany {Udo.Hoenig,
More informationCS 730/730W/830: Intro AI
CS 730/730W/830: Intro AI 1 handout: slides asst 1 milestone was due Wheeler Ruml (UNH) Lecture 4, CS 730 1 / 19 EOLQs Wheeler Ruml (UNH) Lecture 4, CS 730 2 / 19 Comparison Heuristics Search Algorithms
More informationHeuristic Evaluation Function. 11. Evaluation. Issues. Admissible Evaluations. Jonathan Schaeffer
Heuristic Evaluation Function. Evaluation Jonathan Schaeffer jonathan@cs.ualberta.ca www.cs.ualberta.ca/~jonathan Most of the magic in a single-agent searcher is in the evaluation function To obtain an
More informationTrial-based Heuristic Tree Search for Finite Horizon MDPs
Trial-based Heuristic Tree Search for Finite Horizon MDPs Thomas Keller University of Freiburg Freiburg, Germany tkeller@informatik.uni-freiburg.de Malte Helmert University of Basel Basel, Switzerland
More informationCPS 170: Artificial Intelligence Search
CPS 170: Artificial Intelligence http://www.cs.duke.edu/courses/spring09/cps170/ Search Instructor: Vincent Conitzer Search We have some actions that can change the state of the world Change resulting
More informationMarkov Decision Processes and Reinforcement Learning
Lecture 14 and Marco Chiarandini Department of Mathematics & Computer Science University of Southern Denmark Slides by Stuart Russell and Peter Norvig Course Overview Introduction Artificial Intelligence
More informationPlanning & Decision-making in Robotics Case Study: Planning for Autonomous Driving
16-782 Planning & Decision-making in Robotics Case Study: Planning for Autonomous Driving Maxim Likhachev Robotics Institute Carnegie Mellon University Typical Planning Architecture for Autonomous Vehicle
More informationPLite.jl. Release 1.0
PLite.jl Release 1.0 October 19, 2015 Contents 1 In Depth Documentation 3 1.1 Installation................................................ 3 1.2 Problem definition............................................
More informationOutline for today s lecture. Informed Search. Informed Search II. Review: Properties of greedy best-first search. Review: Greedy best-first search:
Outline for today s lecture Informed Search II Informed Search Optimal informed search: A* (AIMA 3.5.2) Creating good heuristic functions Hill Climbing 2 Review: Greedy best-first search: f(n): estimated
More information