Artificial Intelligence
Artificial Intelligence Comprehensive Notes (Sample)
Based around the lecture notes and explanations for the University of Melbourne subject Artificial Intelligence (COMP30024 – Semester 1, 2017) and various online resources. Also incorporates lots of background knowledge and diagrams from the prescribed textbook: Artificial Intelligence: A Modern Approach, by Stuart Russell and Peter Norvig, Pearson, ISBN 9781292024202.
This compilation of notes over the subject is more informative than the lecture slides, but does not go into as much details (which is often unnecessary subject-wise) as the textbook.
Table of Contents (only pages 2-7 in Sample) TABLE OF CONTENTS.............................................................................................................................................. 2 CHAPTER 1: INTELLIGENT AGENTS ......................................................................................................................... 6 THINKING HUMANLY: THE COGNITIVE MODELLING APPROACH .................................................................................................6 ACTING HUMANLY: THE TURING TEST APPROACH .................................................................................................................7 THINKING RATIONALLY: THE “LAWS OF THOUGHT” APPROACH ................................................................................................7 ACTING RATIONALLY: THE RATIONAL AGENT APPROACH .........................................................................................................7 THE AGENT MODEL .......................................................................................................... ERROR! BOOKMARK NOT DEFINED. AGENTS AS A FUNCTION..................................................................................................... ERROR! BOOKMARK NOT DEFINED. AGENT TYPES .................................................................................................................. ERROR! BOOKMARK NOT DEFINED. SIMPLE REFLEX AGENT ....................................................................................................... ERROR! BOOKMARK NOT DEFINED. MODEL-BASED REFLEX AGENT ............................................................................................. ERROR! BOOKMARK NOT DEFINED. GOAL-BASED AGENTS ........................................................................................................ ERROR! BOOKMARK NOT DEFINED. UTILITY-BASED AGENT ....................................................................................................... ERROR! BOOKMARK NOT DEFINED. GOALS VS UTILITY-BASED AGENTS........................................................................................ ERROR! BOOKMARK NOT DEFINED. ENVIRONMENT TYPES ....................................................................................................... ERROR! BOOKMARK NOT DEFINED. CHAPTER 2: PROBLEM-SOLVING AGENTS........................................................... ERROR! BOOKMARK NOT DEFINED. SINGLE-STATE PROBLEM FORMULATION ................................................................................ ERROR! BOOKMARK NOT DEFINED. SEARCH ALGORITHMS ....................................................................................................... ERROR! BOOKMARK NOT DEFINED. SEARCH STRATEGIES ......................................................................................................... ERROR! BOOKMARK NOT DEFINED. UNINFORMED SEARCH STRATEGIES ...................................................................................... ERROR! BOOKMARK NOT DEFINED. BREADTH-FIRST SEARCH .................................................................................................... ERROR! BOOKMARK NOT DEFINED. UNIFORM-COST SEARCH (AKA DIJKSTRA’S ALGORITHM)........................................................... ERROR! BOOKMARK NOT DEFINED. DEPTH-FIRST SEARCH ........................................................................................................ ERROR! BOOKMARK NOT DEFINED. DEPTH-LIMITED SEARCH .................................................................................................... ERROR! BOOKMARK NOT DEFINED. ITERATIVE DEEPENING SEARCH............................................................................................ ERROR! BOOKMARK NOT DEFINED. BIDIRECTIONAL SEARCH ..................................................................................................... ERROR! BOOKMARK NOT DEFINED. COMPARISON OF DIFFERENT UNINFORMED SEARCH STRATEGIES................................................. ERROR! BOOKMARK NOT DEFINED. CHAPTER 3: INFORMED SEARCH STRATEGIES .................................................... ERROR! BOOKMARK NOT DEFINED. BEST-FIRST SEARCH ........................................................................................................... ERROR! BOOKMARK NOT DEFINED. GREEDY BEST-FIRST SEARCH................................................................................................ ERROR! BOOKMARK NOT DEFINED. A* SEARCH ..................................................................................................................... ERROR! BOOKMARK NOT DEFINED. ADMISSIBILITY ................................................................................................................. ERROR! BOOKMARK NOT DEFINED. PROPERTIES OF A* SEARCH ................................................................................................ ERROR! BOOKMARK NOT DEFINED. DOMINANCE (HEURISTICS COMPARISON) .............................................................................. ERROR! BOOKMARK NOT DEFINED. RELAXED PROBLEMS ......................................................................................................... ERROR! BOOKMARK NOT DEFINED. ITERATIVE IMPROVEMENT ALGORITHMS ................................................................................ ERROR! BOOKMARK NOT DEFINED. HILL-CLIMBING SEARCH ..................................................................................................... ERROR! BOOKMARK NOT DEFINED.
CHAPTER 4: GAME PLAYING AND ADVERSARIAL SEARCH .................................. ERROR! BOOKMARK NOT DEFINED. MULTI-AGENT ENVIRONMENT............................................................................................. ERROR! BOOKMARK NOT DEFINED. TYPES OF GAMES .............................................................................................................. ERROR! BOOKMARK NOT DEFINED. REPRESENTING A GAME AS A SEARCH PROBLEM ...................................................................... ERROR! BOOKMARK NOT DEFINED. MINIMAX ....................................................................................................................... ERROR! BOOKMARK NOT DEFINED. PROPERTIES OF MINIMAX .................................................................................................. ERROR! BOOKMARK NOT DEFINED. RESOURCE LIMITS ............................................................................................................. ERROR! BOOKMARK NOT DEFINED. EVALUATION FUNCTIONS ................................................................................................... ERROR! BOOKMARK NOT DEFINED. EXACT VALUES DON’T MATTER (DETERMINISTIC) ..................................................................... ERROR! BOOKMARK NOT DEFINED. Α–Β PRUNING .................................................................................................................. ERROR! BOOKMARK NOT DEFINED. NON-DETERMINISTIC GAMES .............................................................................................. ERROR! BOOKMARK NOT DEFINED. EXACT VALUES DO MATTER (NON-DETERMINISTIC) .................................................................. ERROR! BOOKMARK NOT DEFINED. CHAPTER 5: MACHINE LEARNING IN GAME SEARCH .......................................... ERROR! BOOKMARK NOT DEFINED. SUPERVISED LEARNING USING GRADIENT DESCENT SEARCH ....................................................... ERROR! BOOKMARK NOT DEFINED. TEMPORAL DIFFERENCE LEARNING IN GAMES ......................................................................... ERROR! BOOKMARK NOT DEFINED. CHALLENGES IN GAME AGENT DESIGN ................................................................................... ERROR! BOOKMARK NOT DEFINED. BOOK LEARNING .............................................................................................................. ERROR! BOOKMARK NOT DEFINED. SEARCH CONTROL LEARNING .............................................................................................. ERROR! BOOKMARK NOT DEFINED. LEARNING EVALUATION FUNCTION WEIGHTS .......................................................................... ERROR! BOOKMARK NOT DEFINED. SUPERVISED LEARNING ...................................................................................................... ERROR! BOOKMARK NOT DEFINED. GRADIENT DESCENT LEARNING ............................................................................................ ERROR! BOOKMARK NOT DEFINED. TEMPORAL DIFFERENCE LEARNING ...................................................................................... ERROR! BOOKMARK NOT DEFINED. TD LEAF (Λ) .................................................................................................................... ERROR! BOOKMARK NOT DEFINED. ENVIRONMENT FOR LEARNING ............................................................................................ ERROR! BOOKMARK NOT DEFINED. CHAPTER 6: CONSTRAINT SATISFACTION PROBLEM .......................................... ERROR! BOOKMARK NOT DEFINED. CONSTRAINT GRAPH ......................................................................................................... ERROR! BOOKMARK NOT DEFINED. WHY FORMULATE A PROBLEM AS A CSP? ............................................................................. ERROR! BOOKMARK NOT DEFINED. VARIATIONS ON THE CSP FORMALISM .................................................................................. ERROR! BOOKMARK NOT DEFINED. WHAT ARE THE APPLICATIONS OF CSPS IN THE REAL WORLD? ................................................... ERROR! BOOKMARK NOT DEFINED. STANDARD SEARCH FORMULATION (NAIVE APPROACH) ........................................................... ERROR! BOOKMARK NOT DEFINED. BACKTRACKING SEARCH ..................................................................................................... ERROR! BOOKMARK NOT DEFINED. VARIABLE AND VALUE ORDERING ......................................................................................... ERROR! BOOKMARK NOT DEFINED. CONSTRAINT PROPAGATION: INFERENCE IN CSPS ................................................................... ERROR! BOOKMARK NOT DEFINED. NODE CONSISTENCY .......................................................................................................... ERROR! BOOKMARK NOT DEFINED. ARC CONSISTENCY ............................................................................................................ ERROR! BOOKMARK NOT DEFINED. INTERLEAVING SEARCH AND INFERENCE ................................................................................. ERROR! BOOKMARK NOT DEFINED. FORWARD CHECKING ........................................................................................................ ERROR! BOOKMARK NOT DEFINED. MAINTAINING ARC CONSISTENCY (NOT COVERED IN SYLLABUS BUT COMPLETES THE TOPIC) ............ ERROR! BOOKMARK NOT DEFINED. THE STRUCTURE OF PROBLEMS ........................................................................................... ERROR! BOOKMARK NOT DEFINED. BENEFITS OF BREAKING A PROBLEM INTO SUBPROBLEMS .......................................................... ERROR! BOOKMARK NOT DEFINED. TREE-STRUCTURED CSPS ................................................................................................... ERROR! BOOKMARK NOT DEFINED. TOPOLOGICAL SORT (THE UN-NAMED ALGORITHM IN THE LECTURE SLIDES).................................. ERROR! BOOKMARK NOT DEFINED. ALMOST-A-TREE SCENARIO ................................................................................................ ERROR! BOOKMARK NOT DEFINED. LOCAL SEARCH FOR CSPS .................................................................................................. ERROR! BOOKMARK NOT DEFINED.
CHAPTER 7: UNCERTAINTY................................................................................. ERROR! BOOKMARK NOT DEFINED. METHODS FOR HANDLING UNCERTAINTY ............................................................................... ERROR! BOOKMARK NOT DEFINED. SUMMARIZING UNCERTAINTY ............................................................................................. ERROR! BOOKMARK NOT DEFINED. SUBJECTIVE OR BAYESIAN PROBABILITY................................................................................. ERROR! BOOKMARK NOT DEFINED. UNCERTAINTY AND RATIONAL DECISIONS............................................................................... ERROR! BOOKMARK NOT DEFINED. PROBABILITY ................................................................................................................... ERROR! BOOKMARK NOT DEFINED. RANDOM VARIABLES......................................................................................................... ERROR! BOOKMARK NOT DEFINED. PROPOSITIONS ................................................................................................................. ERROR! BOOKMARK NOT DEFINED. THE LANGUAGE OF PROPOSITION IN PROBABILITY ASSERTIONS ................................................... ERROR! BOOKMARK NOT DEFINED. UNCONDITIONAL PROBABILITIES.......................................................................................... ERROR! BOOKMARK NOT DEFINED. PROBABILITY DISTRIBUTION................................................................................................ ERROR! BOOKMARK NOT DEFINED. JOINT PROBABILITY DISTRIBUTION ....................................................................................... ERROR! BOOKMARK NOT DEFINED. CONDITIONAL PROBABILITY ................................................................................................ ERROR! BOOKMARK NOT DEFINED. INFERENCE USING FULL JOINT DISTRIBUTIONS ........................................................................ ERROR! BOOKMARK NOT DEFINED. NORMALIZATION.............................................................................................................. ERROR! BOOKMARK NOT DEFINED. GENERAL INFERENCE PROCEDURE........................................................................................ ERROR! BOOKMARK NOT DEFINED. INDEPENDENCE ................................................................................................................ ERROR! BOOKMARK NOT DEFINED. BAYES’ RULE AND ITS USE ................................................................................................... ERROR! BOOKMARK NOT DEFINED. APPLYING BAYES’ RULE...................................................................................................... ERROR! BOOKMARK NOT DEFINED. COMBINING EVIDENCE USING BAYES’ RULE ............................................................................ ERROR! BOOKMARK NOT DEFINED. CHAPTER 8: BAYESIAN BELIEF NETWORKS ......................................................... ERROR! BOOKMARK NOT DEFINED. SYNTAX .......................................................................................................................... ERROR! BOOKMARK NOT DEFINED. SEMANTICS ..................................................................................................................... ERROR! BOOKMARK NOT DEFINED. REPRESENTING THE FULL JOINT DISTRIBUTION ........................................................................ ERROR! BOOKMARK NOT DEFINED. CONSTRUCTING BAYESIAN NETWORKS ................................................................................. ERROR! BOOKMARK NOT DEFINED. COMPACTNESS ................................................................................................................ ERROR! BOOKMARK NOT DEFINED. NODE ORDERING ............................................................................................................. ERROR! BOOKMARK NOT DEFINED. INFERENCE IN BAYESIAN NETWORKS .................................................................................... ERROR! BOOKMARK NOT DEFINED. INFERENCE BY ENUMERATION ............................................................................................. ERROR! BOOKMARK NOT DEFINED. CHAPTER 9: MAKING COMPLEX DECISIONS – AUCTIONS ................................... ERROR! BOOKMARK NOT DEFINED. MECHANISM DESIGN ........................................................................................................ ERROR! BOOKMARK NOT DEFINED. AUCTIONS....................................................................................................................... ERROR! BOOKMARK NOT DEFINED. DESIRABLE PROPERTIES OF AN AUCTION ................................................................................ ERROR! BOOKMARK NOT DEFINED. DIMENSIONS OF AUCTION PROTOCOLS .................................................................................. ERROR! BOOKMARK NOT DEFINED. ENGLISH AUCTION ............................................................................................................ ERROR! BOOKMARK NOT DEFINED. DUTCH AUCTION .............................................................................................................. ERROR! BOOKMARK NOT DEFINED. FIRST-PRICE, SEALED-BID AUCTION ....................................................................................... ERROR! BOOKMARK NOT DEFINED. SECOND-PRICE, SEALED-BID AUCTION (VICKREY AUCTION) ........................................................ ERROR! BOOKMARK NOT DEFINED. CHAPTER 10: ROBOTICS ..................................................................................... ERROR! BOOKMARK NOT DEFINED. EFFECTORS AND SENSORS .................................................................................................. ERROR! BOOKMARK NOT DEFINED. MANIPULATORS ............................................................................................................... ERROR! BOOKMARK NOT DEFINED. NON-HOLONOMIC ROBOTS ................................................................................................ ERROR! BOOKMARK NOT DEFINED. ROBOTIC UNCERTAINTY ..................................................................................................... ERROR! BOOKMARK NOT DEFINED. SOURCES OF UNCERTAINTY IN INTERACTION ........................................................................... ERROR! BOOKMARK NOT DEFINED. SOURCES OF UNCERTAINTY IN MOTION ACTIONS ..................................................................... ERROR! BOOKMARK NOT DEFINED. CLASSIFICATIONS OF STATES ............................................................................................... ERROR! BOOKMARK NOT DEFINED.
DEALING WITH UNCERTAINTY.............................................................................................. ERROR! BOOKMARK NOT DEFINED. IMPLICATIONS OF SENSOR UNCERTAINTY ............................................................................... ERROR! BOOKMARK NOT DEFINED. ROBOTIC PERCEPTION ....................................................................................................... ERROR! BOOKMARK NOT DEFINED. LOCALIZATION ................................................................................................................. ERROR! BOOKMARK NOT DEFINED. PARTICLE FILTERING .......................................................................................................... ERROR! BOOKMARK NOT DEFINED. LANDMARKS .................................................................................................................... ERROR! BOOKMARK NOT DEFINED. MAPPING ....................................................................................................................... ERROR! BOOKMARK NOT DEFINED. SIMULTANEOUS LOCALIZATION AND MAPPING (SLAM) .......................................................... ERROR! BOOKMARK NOT DEFINED. BAYESIAN INFERENCE ON SENSORS ....................................................................................... ERROR! BOOKMARK NOT DEFINED. BAYES LAW: REVISION ...................................................................................................... ERROR! BOOKMARK NOT DEFINED. SENSOR DETECTION RATES.................................................................................................. ERROR! BOOKMARK NOT DEFINED. MOTION PLANNING .......................................................................................................... ERROR! BOOKMARK NOT DEFINED. CELL DECOMPOSITION ....................................................................................................... ERROR! BOOKMARK NOT DEFINED. SKELETONIZATION ............................................................................................................ ERROR! BOOKMARK NOT DEFINED. SKELETONIZATION: VORONOI GRAPH.................................................................................... ERROR! BOOKMARK NOT DEFINED. SKELETONIZATION: PROBABILISTIC ROADMAP ........................................................................ ERROR! BOOKMARK NOT DEFINED.
Chapter 1: Intelligent Agents An agent is anything that can be viewed as perceiving its environment through sensors and acting upon that environment through actuators. For example, a human agent has eyes, ears and other organs for sensors; and hands, legs, vocal tract and so on as actuators. A robotic agent might have cameras and infrared range finders as sensors and various motors as actuators.
What does it mean to do the right thing? By considering the consequences of the agent’s behavior. An agent generates a sequence of actions based on the percepts it receives in a certain environment. This sequence of actions causes the environment to go through a sequence of states. If the sequence is desirable, then the agent has performed well. This notion of desirability is captured by a performance measure that evaluates any given sequence of environment states. As a general rule, it is better to design performance measures according to what one actually wants in the environment, rather than to how one thinks the agent should behave.
Rationality is an example of an ideal performance measure. We adopt the view that intelligence is concerned mainly with rational action. Ideally, an intelligent agent takes the best possible action in a situation. We define a rational agent as follows: For each possible percept sequence, a rational agent should select an action that is expected to maximize its performance measure, given the evidence provided by the percept sequence and whatever built-in knowledge the agent has.
We define AI as the study of agents that receive percepts from the environment and perform actions.
Thinking humanly: The cognitive modelling approach We must have some way of determining how humans think. We need to get inside the actual workings of human minds. There are three ways to do this: o o o
Through introspection – trying to catch our own thoughts as they go by; Through psychological experiments – observing a person in action; And through brain imaging – observing the brain in action.
Once we have a sufficiently precise theory of the mind, it becomes possible to express the theory as a computer program. If the program’s input-output behavior matches corresponding human behavior, that is evidence that some of the program’s mechanisms could also be operating in humans. We should be more concerned with comparing the trace of the programs reasoning steps to traces of human subjects solving the same problems, as opposed to just having the program solve problems correctly.
Acting humanly: The Turing Test approach The computer would need to possess the following capabilities: o o o o
Knowledge representation to store what it knows or hears; Automated reasoning to use the stored information to answer questions and to draw new conclusions; Natural language processing to enable it to communicate successfully in English; Machine learning to adapt to new circumstances and to detect and extrapolate patterns.
The quest for “artificial flight” succeeded when the Wright brothers and others stopped imitating birds and started using wind tunnels and learning about aerodynamics. Aeronautical engineering texts do not define the goal of their field as making “machines that fly so exactly like pigeons that they can fool even other pigeons”.
Thinking rationally: The “laws of thought” approach Aristotle attempted to codify “right thinking,” that is, irrefutable reasoning process. His syllogisms provided patterns for argument structures that always yielded correct conclusions when given correct premises. There are two main obstacles to this approach. First, it is not easy to take informal knowledge and state it in the formal terms required by logical notation, particularly with a degree of knowledge uncertainty. Second, there is a substantial difference between solving a problem “in principle” and solving it in practice.
Acting rationally: The rational agent approach Computer agents are expected to do more than normal programs: operate autonomously, perceive their environment, persist over a prolonged time period, adapt to change, and create and pursue goals. A rational agent is one that acts so as to achieve the best (expected) outcome. In the “laws of thought” approach to AI, the emphasis was on correct inferences. Making correct inferences is sometimes part of being a rational agent, because one way to act rationally is to reason logically to the conclusion that a given action will achieve one’s goals and then to act on that conclusion. On the other hand, correct inference is not all of rationality; in some situations, there is no provably correct thing to do, but something must still be done. There are also ways of acting rationally that cannot be said to involve inference. For example, recoiling from a hot stove is a reflex action that is usually more successful than a slower action taken after careful deliberation.
Based around the lecture notes and explanations for the University of Melbourne subject Artificial Intelligence (COMP30024 – Semester 1, 2017) and various online resources. Also incorporates lots of background knowledge and diagrams from the prescribed textbook: Artificial Intelligence: A Modern Approach, by Stuart Russell and Peter Norvig, Pearson, ISBN 9781292024202.
This compilation of notes over the subject is more informative than the lecture slides, but does not go into as much details (which is often unnecessary subject-wise) as the textbook.
Table of Contents (only pages 2-7 in Sample) TABLE OF CONTENTS.............................................................................................................................................. 2 CHAPTER 1: INTELLIGENT AGENTS ......................................................................................................................... 6 THINKING HUMANLY: THE COGNITIVE MODELLING APPROACH .................................................................................................6 ACTING HUMANLY: THE TURING TEST APPROACH .................................................................................................................7 THINKING RATIONALLY: THE “LAWS OF THOUGHT” APPROACH ................................................................................................7 ACTING RATIONALLY: THE RATIONAL AGENT APPROACH .........................................................................................................7 THE AGENT MODEL .......................................................................................................... ERROR! BOOKMARK NOT DEFINED. AGENTS AS A FUNCTION..................................................................................................... ERROR! BOOKMARK NOT DEFINED. AGENT TYPES .................................................................................................................. ERROR! BOOKMARK NOT DEFINED. SIMPLE REFLEX AGENT ....................................................................................................... ERROR! BOOKMARK NOT DEFINED. MODEL-BASED REFLEX AGENT ............................................................................................. ERROR! BOOKMARK NOT DEFINED. GOAL-BASED AGENTS ........................................................................................................ ERROR! BOOKMARK NOT DEFINED. UTILITY-BASED AGENT ....................................................................................................... ERROR! BOOKMARK NOT DEFINED. GOALS VS UTILITY-BASED AGENTS........................................................................................ ERROR! BOOKMARK NOT DEFINED. ENVIRONMENT TYPES ....................................................................................................... ERROR! BOOKMARK NOT DEFINED. CHAPTER 2: PROBLEM-SOLVING AGENTS........................................................... ERROR! BOOKMARK NOT DEFINED. SINGLE-STATE PROBLEM FORMULATION ................................................................................ ERROR! BOOKMARK NOT DEFINED. SEARCH ALGORITHMS ....................................................................................................... ERROR! BOOKMARK NOT DEFINED. SEARCH STRATEGIES ......................................................................................................... ERROR! BOOKMARK NOT DEFINED. UNINFORMED SEARCH STRATEGIES ...................................................................................... ERROR! BOOKMARK NOT DEFINED. BREADTH-FIRST SEARCH .................................................................................................... ERROR! BOOKMARK NOT DEFINED. UNIFORM-COST SEARCH (AKA DIJKSTRA’S ALGORITHM)........................................................... ERROR! BOOKMARK NOT DEFINED. DEPTH-FIRST SEARCH ........................................................................................................ ERROR! BOOKMARK NOT DEFINED. DEPTH-LIMITED SEARCH .................................................................................................... ERROR! BOOKMARK NOT DEFINED. ITERATIVE DEEPENING SEARCH............................................................................................ ERROR! BOOKMARK NOT DEFINED. BIDIRECTIONAL SEARCH ..................................................................................................... ERROR! BOOKMARK NOT DEFINED. COMPARISON OF DIFFERENT UNINFORMED SEARCH STRATEGIES................................................. ERROR! BOOKMARK NOT DEFINED. CHAPTER 3: INFORMED SEARCH STRATEGIES .................................................... ERROR! BOOKMARK NOT DEFINED. BEST-FIRST SEARCH ........................................................................................................... ERROR! BOOKMARK NOT DEFINED. GREEDY BEST-FIRST SEARCH................................................................................................ ERROR! BOOKMARK NOT DEFINED. A* SEARCH ..................................................................................................................... ERROR! BOOKMARK NOT DEFINED. ADMISSIBILITY ................................................................................................................. ERROR! BOOKMARK NOT DEFINED. PROPERTIES OF A* SEARCH ................................................................................................ ERROR! BOOKMARK NOT DEFINED. DOMINANCE (HEURISTICS COMPARISON) .............................................................................. ERROR! BOOKMARK NOT DEFINED. RELAXED PROBLEMS ......................................................................................................... ERROR! BOOKMARK NOT DEFINED. ITERATIVE IMPROVEMENT ALGORITHMS ................................................................................ ERROR! BOOKMARK NOT DEFINED. HILL-CLIMBING SEARCH ..................................................................................................... ERROR! BOOKMARK NOT DEFINED.
CHAPTER 4: GAME PLAYING AND ADVERSARIAL SEARCH .................................. ERROR! BOOKMARK NOT DEFINED. MULTI-AGENT ENVIRONMENT............................................................................................. ERROR! BOOKMARK NOT DEFINED. TYPES OF GAMES .............................................................................................................. ERROR! BOOKMARK NOT DEFINED. REPRESENTING A GAME AS A SEARCH PROBLEM ...................................................................... ERROR! BOOKMARK NOT DEFINED. MINIMAX ....................................................................................................................... ERROR! BOOKMARK NOT DEFINED. PROPERTIES OF MINIMAX .................................................................................................. ERROR! BOOKMARK NOT DEFINED. RESOURCE LIMITS ............................................................................................................. ERROR! BOOKMARK NOT DEFINED. EVALUATION FUNCTIONS ................................................................................................... ERROR! BOOKMARK NOT DEFINED. EXACT VALUES DON’T MATTER (DETERMINISTIC) ..................................................................... ERROR! BOOKMARK NOT DEFINED. Α–Β PRUNING .................................................................................................................. ERROR! BOOKMARK NOT DEFINED. NON-DETERMINISTIC GAMES .............................................................................................. ERROR! BOOKMARK NOT DEFINED. EXACT VALUES DO MATTER (NON-DETERMINISTIC) .................................................................. ERROR! BOOKMARK NOT DEFINED. CHAPTER 5: MACHINE LEARNING IN GAME SEARCH .......................................... ERROR! BOOKMARK NOT DEFINED. SUPERVISED LEARNING USING GRADIENT DESCENT SEARCH ....................................................... ERROR! BOOKMARK NOT DEFINED. TEMPORAL DIFFERENCE LEARNING IN GAMES ......................................................................... ERROR! BOOKMARK NOT DEFINED. CHALLENGES IN GAME AGENT DESIGN ................................................................................... ERROR! BOOKMARK NOT DEFINED. BOOK LEARNING .............................................................................................................. ERROR! BOOKMARK NOT DEFINED. SEARCH CONTROL LEARNING .............................................................................................. ERROR! BOOKMARK NOT DEFINED. LEARNING EVALUATION FUNCTION WEIGHTS .......................................................................... ERROR! BOOKMARK NOT DEFINED. SUPERVISED LEARNING ...................................................................................................... ERROR! BOOKMARK NOT DEFINED. GRADIENT DESCENT LEARNING ............................................................................................ ERROR! BOOKMARK NOT DEFINED. TEMPORAL DIFFERENCE LEARNING ...................................................................................... ERROR! BOOKMARK NOT DEFINED. TD LEAF (Λ) .................................................................................................................... ERROR! BOOKMARK NOT DEFINED. ENVIRONMENT FOR LEARNING ............................................................................................ ERROR! BOOKMARK NOT DEFINED. CHAPTER 6: CONSTRAINT SATISFACTION PROBLEM .......................................... ERROR! BOOKMARK NOT DEFINED. CONSTRAINT GRAPH ......................................................................................................... ERROR! BOOKMARK NOT DEFINED. WHY FORMULATE A PROBLEM AS A CSP? ............................................................................. ERROR! BOOKMARK NOT DEFINED. VARIATIONS ON THE CSP FORMALISM .................................................................................. ERROR! BOOKMARK NOT DEFINED. WHAT ARE THE APPLICATIONS OF CSPS IN THE REAL WORLD? ................................................... ERROR! BOOKMARK NOT DEFINED. STANDARD SEARCH FORMULATION (NAIVE APPROACH) ........................................................... ERROR! BOOKMARK NOT DEFINED. BACKTRACKING SEARCH ..................................................................................................... ERROR! BOOKMARK NOT DEFINED. VARIABLE AND VALUE ORDERING ......................................................................................... ERROR! BOOKMARK NOT DEFINED. CONSTRAINT PROPAGATION: INFERENCE IN CSPS ................................................................... ERROR! BOOKMARK NOT DEFINED. NODE CONSISTENCY .......................................................................................................... ERROR! BOOKMARK NOT DEFINED. ARC CONSISTENCY ............................................................................................................ ERROR! BOOKMARK NOT DEFINED. INTERLEAVING SEARCH AND INFERENCE ................................................................................. ERROR! BOOKMARK NOT DEFINED. FORWARD CHECKING ........................................................................................................ ERROR! BOOKMARK NOT DEFINED. MAINTAINING ARC CONSISTENCY (NOT COVERED IN SYLLABUS BUT COMPLETES THE TOPIC) ............ ERROR! BOOKMARK NOT DEFINED. THE STRUCTURE OF PROBLEMS ........................................................................................... ERROR! BOOKMARK NOT DEFINED. BENEFITS OF BREAKING A PROBLEM INTO SUBPROBLEMS .......................................................... ERROR! BOOKMARK NOT DEFINED. TREE-STRUCTURED CSPS ................................................................................................... ERROR! BOOKMARK NOT DEFINED. TOPOLOGICAL SORT (THE UN-NAMED ALGORITHM IN THE LECTURE SLIDES).................................. ERROR! BOOKMARK NOT DEFINED. ALMOST-A-TREE SCENARIO ................................................................................................ ERROR! BOOKMARK NOT DEFINED. LOCAL SEARCH FOR CSPS .................................................................................................. ERROR! BOOKMARK NOT DEFINED.
CHAPTER 7: UNCERTAINTY................................................................................. ERROR! BOOKMARK NOT DEFINED. METHODS FOR HANDLING UNCERTAINTY ............................................................................... ERROR! BOOKMARK NOT DEFINED. SUMMARIZING UNCERTAINTY ............................................................................................. ERROR! BOOKMARK NOT DEFINED. SUBJECTIVE OR BAYESIAN PROBABILITY................................................................................. ERROR! BOOKMARK NOT DEFINED. UNCERTAINTY AND RATIONAL DECISIONS............................................................................... ERROR! BOOKMARK NOT DEFINED. PROBABILITY ................................................................................................................... ERROR! BOOKMARK NOT DEFINED. RANDOM VARIABLES......................................................................................................... ERROR! BOOKMARK NOT DEFINED. PROPOSITIONS ................................................................................................................. ERROR! BOOKMARK NOT DEFINED. THE LANGUAGE OF PROPOSITION IN PROBABILITY ASSERTIONS ................................................... ERROR! BOOKMARK NOT DEFINED. UNCONDITIONAL PROBABILITIES.......................................................................................... ERROR! BOOKMARK NOT DEFINED. PROBABILITY DISTRIBUTION................................................................................................ ERROR! BOOKMARK NOT DEFINED. JOINT PROBABILITY DISTRIBUTION ....................................................................................... ERROR! BOOKMARK NOT DEFINED. CONDITIONAL PROBABILITY ................................................................................................ ERROR! BOOKMARK NOT DEFINED. INFERENCE USING FULL JOINT DISTRIBUTIONS ........................................................................ ERROR! BOOKMARK NOT DEFINED. NORMALIZATION.............................................................................................................. ERROR! BOOKMARK NOT DEFINED. GENERAL INFERENCE PROCEDURE........................................................................................ ERROR! BOOKMARK NOT DEFINED. INDEPENDENCE ................................................................................................................ ERROR! BOOKMARK NOT DEFINED. BAYES’ RULE AND ITS USE ................................................................................................... ERROR! BOOKMARK NOT DEFINED. APPLYING BAYES’ RULE...................................................................................................... ERROR! BOOKMARK NOT DEFINED. COMBINING EVIDENCE USING BAYES’ RULE ............................................................................ ERROR! BOOKMARK NOT DEFINED. CHAPTER 8: BAYESIAN BELIEF NETWORKS ......................................................... ERROR! BOOKMARK NOT DEFINED. SYNTAX .......................................................................................................................... ERROR! BOOKMARK NOT DEFINED. SEMANTICS ..................................................................................................................... ERROR! BOOKMARK NOT DEFINED. REPRESENTING THE FULL JOINT DISTRIBUTION ........................................................................ ERROR! BOOKMARK NOT DEFINED. CONSTRUCTING BAYESIAN NETWORKS ................................................................................. ERROR! BOOKMARK NOT DEFINED. COMPACTNESS ................................................................................................................ ERROR! BOOKMARK NOT DEFINED. NODE ORDERING ............................................................................................................. ERROR! BOOKMARK NOT DEFINED. INFERENCE IN BAYESIAN NETWORKS .................................................................................... ERROR! BOOKMARK NOT DEFINED. INFERENCE BY ENUMERATION ............................................................................................. ERROR! BOOKMARK NOT DEFINED. CHAPTER 9: MAKING COMPLEX DECISIONS – AUCTIONS ................................... ERROR! BOOKMARK NOT DEFINED. MECHANISM DESIGN ........................................................................................................ ERROR! BOOKMARK NOT DEFINED. AUCTIONS....................................................................................................................... ERROR! BOOKMARK NOT DEFINED. DESIRABLE PROPERTIES OF AN AUCTION ................................................................................ ERROR! BOOKMARK NOT DEFINED. DIMENSIONS OF AUCTION PROTOCOLS .................................................................................. ERROR! BOOKMARK NOT DEFINED. ENGLISH AUCTION ............................................................................................................ ERROR! BOOKMARK NOT DEFINED. DUTCH AUCTION .............................................................................................................. ERROR! BOOKMARK NOT DEFINED. FIRST-PRICE, SEALED-BID AUCTION ....................................................................................... ERROR! BOOKMARK NOT DEFINED. SECOND-PRICE, SEALED-BID AUCTION (VICKREY AUCTION) ........................................................ ERROR! BOOKMARK NOT DEFINED. CHAPTER 10: ROBOTICS ..................................................................................... ERROR! BOOKMARK NOT DEFINED. EFFECTORS AND SENSORS .................................................................................................. ERROR! BOOKMARK NOT DEFINED. MANIPULATORS ............................................................................................................... ERROR! BOOKMARK NOT DEFINED. NON-HOLONOMIC ROBOTS ................................................................................................ ERROR! BOOKMARK NOT DEFINED. ROBOTIC UNCERTAINTY ..................................................................................................... ERROR! BOOKMARK NOT DEFINED. SOURCES OF UNCERTAINTY IN INTERACTION ........................................................................... ERROR! BOOKMARK NOT DEFINED. SOURCES OF UNCERTAINTY IN MOTION ACTIONS ..................................................................... ERROR! BOOKMARK NOT DEFINED. CLASSIFICATIONS OF STATES ............................................................................................... ERROR! BOOKMARK NOT DEFINED.
DEALING WITH UNCERTAINTY.............................................................................................. ERROR! BOOKMARK NOT DEFINED. IMPLICATIONS OF SENSOR UNCERTAINTY ............................................................................... ERROR! BOOKMARK NOT DEFINED. ROBOTIC PERCEPTION ....................................................................................................... ERROR! BOOKMARK NOT DEFINED. LOCALIZATION ................................................................................................................. ERROR! BOOKMARK NOT DEFINED. PARTICLE FILTERING .......................................................................................................... ERROR! BOOKMARK NOT DEFINED. LANDMARKS .................................................................................................................... ERROR! BOOKMARK NOT DEFINED. MAPPING ....................................................................................................................... ERROR! BOOKMARK NOT DEFINED. SIMULTANEOUS LOCALIZATION AND MAPPING (SLAM) .......................................................... ERROR! BOOKMARK NOT DEFINED. BAYESIAN INFERENCE ON SENSORS ....................................................................................... ERROR! BOOKMARK NOT DEFINED. BAYES LAW: REVISION ...................................................................................................... ERROR! BOOKMARK NOT DEFINED. SENSOR DETECTION RATES.................................................................................................. ERROR! BOOKMARK NOT DEFINED. MOTION PLANNING .......................................................................................................... ERROR! BOOKMARK NOT DEFINED. CELL DECOMPOSITION ....................................................................................................... ERROR! BOOKMARK NOT DEFINED. SKELETONIZATION ............................................................................................................ ERROR! BOOKMARK NOT DEFINED. SKELETONIZATION: VORONOI GRAPH.................................................................................... ERROR! BOOKMARK NOT DEFINED. SKELETONIZATION: PROBABILISTIC ROADMAP ........................................................................ ERROR! BOOKMARK NOT DEFINED.
Chapter 1: Intelligent Agents An agent is anything that can be viewed as perceiving its environment through sensors and acting upon that environment through actuators. For example, a human agent has eyes, ears and other organs for sensors; and hands, legs, vocal tract and so on as actuators. A robotic agent might have cameras and infrared range finders as sensors and various motors as actuators.
What does it mean to do the right thing? By considering the consequences of the agent’s behavior. An agent generates a sequence of actions based on the percepts it receives in a certain environment. This sequence of actions causes the environment to go through a sequence of states. If the sequence is desirable, then the agent has performed well. This notion of desirability is captured by a performance measure that evaluates any given sequence of environment states. As a general rule, it is better to design performance measures according to what one actually wants in the environment, rather than to how one thinks the agent should behave.
Rationality is an example of an ideal performance measure. We adopt the view that intelligence is concerned mainly with rational action. Ideally, an intelligent agent takes the best possible action in a situation. We define a rational agent as follows: For each possible percept sequence, a rational agent should select an action that is expected to maximize its performance measure, given the evidence provided by the percept sequence and whatever built-in knowledge the agent has.
We define AI as the study of agents that receive percepts from the environment and perform actions.
Thinking humanly: The cognitive modelling approach We must have some way of determining how humans think. We need to get inside the actual workings of human minds. There are three ways to do this: o o o
Through introspection – trying to catch our own thoughts as they go by; Through psychological experiments – observing a person in action; And through brain imaging – observing the brain in action.
Once we have a sufficiently precise theory of the mind, it becomes possible to express the theory as a computer program. If the program’s input-output behavior matches corresponding human behavior, that is evidence that some of the program’s mechanisms could also be operating in humans. We should be more concerned with comparing the trace of the programs reasoning steps to traces of human subjects solving the same problems, as opposed to just having the program solve problems correctly.
Acting humanly: The Turing Test approach The computer would need to possess the following capabilities: o o o o
Knowledge representation to store what it knows or hears; Automated reasoning to use the stored information to answer questions and to draw new conclusions; Natural language processing to enable it to communicate successfully in English; Machine learning to adapt to new circumstances and to detect and extrapolate patterns.
The quest for “artificial flight” succeeded when the Wright brothers and others stopped imitating birds and started using wind tunnels and learning about aerodynamics. Aeronautical engineering texts do not define the goal of their field as making “machines that fly so exactly like pigeons that they can fool even other pigeons”.
Thinking rationally: The “laws of thought” approach Aristotle attempted to codify “right thinking,” that is, irrefutable reasoning process. His syllogisms provided patterns for argument structures that always yielded correct conclusions when given correct premises. There are two main obstacles to this approach. First, it is not easy to take informal knowledge and state it in the formal terms required by logical notation, particularly with a degree of knowledge uncertainty. Second, there is a substantial difference between solving a problem “in principle” and solving it in practice.
Acting rationally: The rational agent approach Computer agents are expected to do more than normal programs: operate autonomously, perceive their environment, persist over a prolonged time period, adapt to change, and create and pursue goals. A rational agent is one that acts so as to achieve the best (expected) outcome. In the “laws of thought” approach to AI, the emphasis was on correct inferences. Making correct inferences is sometimes part of being a rational agent, because one way to act rationally is to reason logically to the conclusion that a given action will achieve one’s goals and then to act on that conclusion. On the other hand, correct inference is not all of rationality; in some situations, there is no provably correct thing to do, but something must still be done. There are also ways of acting rationally that cannot be said to involve inference. For example, recoiling from a hot stove is a reflex action that is usually more successful than a slower action taken after careful deliberation.
