Bob Marinier Advisor: John Laird Functional Contributions of Emotion to Artificial Intelligence.

Slides:



Advertisements
Similar presentations
Pat Langley School of Computing and Informatics Arizona State University Tempe, Arizona USA Modeling Social Cognition in a Unified Cognitive Architecture.
Advertisements

Pat Langley Computational Learning Laboratory Center for the Study of Language and Information Stanford University, Stanford, California USA
Pat Langley Computational Learning Laboratory Center for the Study of Language and Information Stanford University, Stanford, California USA
Pat Langley Institute for the Study of Learning and Expertise Palo Alto, California A Cognitive Architecture for Complex Learning.
Bob Marinier John Laird University of Michigan Electrical Engineering and Computer Science August 2, 2007.
Cognitive Systems, ICANN panel, Q1 What is machine intelligence, as beyond pattern matching, classification and prediction. What is machine intelligence,
A Cognitive Architecture Theory of Comprehension and Appraisal: Unifying Cognitive Functions and Appraisal Bob Marinier John Laird University of Michigan.
1 Soar Emote Bob Marinier John Laird University of Michigan.
Emotion-Driven Reinforcement Learning Bob Marinier & John Laird University of Michigan, Computer Science and Engineering CogSci’08.
Applied Marketing Strategies
Generating Appraisals with Sequence & Influence Networks Bob Marinier, SoarTech 29 th Soar Workshop June 2009.
Intelligent Agents Russell and Norvig: 2
A computational unification of cognitive behavior and emotion Robert P. Marinier III, John E. Laird, Richard L. Lewis Cognitive Systems Research vol. 10,
Bob Marinier Oral Defense University of Michigan, CSE June 17, 2008 A Computational Unification of Cognitive Control, Emotion, and Learning.
Psychology of Learning: Operant Conditioning
A Theory of Unifying Cognitive Processing, Appraisal and Emotion Bob Marinier John Laird University of Michigan.
1 Reinforcement Learning Problem Week #3. Figure reproduced from the figure on page 52 in reference [1] 2 Reinforcement Learning Loop state Agent Environment.
On knowing how to do things: a theory of motor imagery John Annett Cognitive Brain Research 3(1996) Presenter: Chu-Chun Cheng Advisor: Ru-Jung Cherng.
(How) Can Appraisal Theory be Formalized at a Meta-level? Joost Broekens, Doug DeGroot LIACS, Leiden University.
Introduction to SOAR Based on “a gentle introduction to soar: an Architecture for Human Cognition” by Jill Fain Lehman, John Laird, Paul Rosenbloom. Presented.
To what extent do biological and cognitive factors interact?
Introduction to Cognitive Science Lecture #1 : INTRODUCTION Joe Lau Philosophy HKU.
The Importance of Architecture for Achieving Human-level AI John Laird University of Michigan June 17, th Soar Workshop
Learning Goals Learn the consumer market and construct model of consumer buyer behavior Know the four factors that influence buyer behavior Understand.
What is Cognitive Science? … is the interdisciplinary study of mind and intelligence, embracing philosophy, psychology, artificial intelligence, neuroscience,
Modeling Emotion Frameworks Useful In Computation and Speech Frank Enos.
User Psychology ITKP103 Human Being and Information System Sacha Helfenstein Lectures ,
Introduction to Affect and Cognition Copyright © 2005 by The McGraw-Hill Companies, Inc. All rights reserved. McGraw-Hill/Irwin Chapter 3.
Economic Attention Networks: Associative Memory and Resource Allocation for General Intelligence Matthew Iklé, Joel Pitt, Ben Goertzel, George Sellman.
© POSbase 2004 He distinguished five different stimulus evaluation checks: Stimulus Evaluation Checks The model of Scherer, (1986):Scherer, (1986): Scherer.
COMPUTATIONAL MODELING OF INTEGRATED COGNITION AND EMOTION Bob MarinierUniversity of Michigan.
Computational Models of Emotion and Cognition Computational Models of Emotion and cognition Christopher L. Dancy, Frank E. Ritter, Keith Berry Jerry Lin,
Introduction to Behavior- Based Robotics Based on the book Behavior- Based Robotics by Ronald C. Arkin.
What Is Learning? Learning is a relatively permanent change in human capabilities that is not a result of growth processes. These capabilities are related.
Science of Intelligent Systems
An Architecture for Empathic Agents. Abstract Architecture Planning + Coping Deliberated Actions Agent in the World Body Speech Facial expressions Effectors.
SLB /04/07 Thinking and Communicating “The Spiritual Life is Thinking!” (R.B. Thieme, Jr.)
Psychology: memory. Overview An understanding of human memory is critical to an appreciation of how users will store and use relevant information when.
Ecological Interface Design
Integrating high- and low-level Expectations in Deliberative Agents Michele Piunti - Institute of.
TRAINING AND DEVELOPMENT CHAPTER 3 Learning and training Presented by MALIHA PARACHA 2579.
EMOTIONAL INTELLIGENCE. 2 Emotional Intelligence at Work.
Comp 15 - Usability and Human Factors
Mehdi Ghayoumi Kent State University Computer Science Department Summer 2015 Exposition on Cyber Infrastructure and Big Data.
Cognitive Science and Biomedical Informatics Department of Computer Sciences ALMAAREFA COLLEGES.
Copyright  2007 McGraw-Hill Australia Pty Ltd PPTs t/a Consumer Behaviour 5e by Quester, Neal, Pettigrew, Grimmer, Davis & Hawkins Slides prepared by.
Cognitive Architectures For Physical Agents Sara Bolduc Smith College CSC 290.
SOAR A cognitive architecture By: Majid Ali Khan.
1 Learning through Interactive Behavior Specifications Tolga Konik CSLI, Stanford University Douglas Pearson Three Penny Software John Laird University.
Ergonomics/Human Integrated Systems (Project 02)
Cognitive Architectures and General Intelligent Systems Pay Langley 2006 Presentation : Suwang Jang.
Human Learning Ryan Glidden.
1 Situation Comprehension and Emotion Bob Marinier University of Michigan June 2005.
WP6 Emotion in Interaction Embodied Conversational Agents WP6 core task: describe an interactive ECA system with capabilities beyond those of present day.
Cognitive Modeling Cogs 4961, Cogs 6967 Psyc 4510 CSCI 4960 Mike Schoelles
Artificial Intelligence
Consumer Markets and Consumer Buyer Behavior
Learning Fast and Slow John E. Laird
Consumer Markets and Consumer Buyer Behavior
Chapter 7 Psychology: Memory.
Artificial Intelligence Chapter 25 Agent Architectures
Chapter 3: The Reinforcement Learning Problem
SOAR as a Cognitive Architecture for Modeling Driver Workload
Consumer Markets and Consumer Buyer Behavior
Symbolic cognitive architectures
Consumer Markets and Consumer Buyer Behavior
Chapter 3: The Reinforcement Learning Problem
Chapter 3: The Reinforcement Learning Problem
Artificial Intelligence Chapter 25 Agent Architectures
How Bodies Matter to Minds
Presentation transcript:

Bob Marinier Advisor: John Laird Functional Contributions of Emotion to Artificial Intelligence

Introduction 2 Folk psychology considers emotions to be a distraction from logical thought People tend to think that emotion is unknowable, indefinable Psychological work in the last several decades has demonstrated that emotion plays a critical role in effective functioning and learning

Introduction 3 Research Goals Bring the functionality of emotion to AI Create a precise computational definition of emotion Approach Integrate emotion with a complete agent framework Computationally distinguish emotion, mood and feeling Weight feeling’s importance by computing its intensity Use feeling as intrinsic reward signal to drive reinforcement learning

Appraisal Theories of Emotion 4 A situation is evaluated along a number of appraisal dimensions, many of which relate the situation to current goals Novelty, goal relevance, goal conduciveness, expectedness, causal agency, etc. Result of appraisals determines emotion The emotion is combined with mood, which is an “average” over recent emotions, to form a feeling, which is actually perceived with some intensity The feeling can then be coped with (via internal or external actions) Situation Goals Appraisal Emotion, Mood, Feeling Coping

Appraisals to Emotions (Scherer 2001) 5 JoyFearAnger SuddennessHigh/mediumHigh UnpredictabilityHigh Intrinsic pleasantnessLow Goal/need relevanceHigh Cause: agentOther/natureOther Cause: motiveChance/intentionalIntentional Outcome probabilityVery highHighVery high Discrepancy from expectationHigh ConducivenessVery highLow ControlHigh PowerVery lowHigh Why these dimensions? What is the functional purpose?

Functions of Emotion 6 Situation summary: Appraisals and emotion provide abstract interpretation Decouples stimulus/response: Can react to interpretation instead of stimulus Attention: Some appraisals help prioritize processing Historical context: Mood provides a context for current interpretations Learning: Feeling may provide an intrinsic reward signal Memory Decision making Action preparation Communication

Outline 7 Integrate emotion with a complete agent framework Computationally distinguish emotion, mood and feeling Weight feeling’s importance by computing its intensity Use feeling as intrinsic reward signal to drive reinforcement learning Discussion & Conclusion Situation Goals Appraisal Emotion, Mood, Feeling Coping

Newell’s Abstract Functional Operations (Newell 1990) 8 Allen Newell defined a set of computational Abstract Functional Operations that are necessary and sufficient for immediate behavior in humans and complete agents PerceiveObtain raw perception EncodeCreate domain-independent representation AttendChoose stimulus to process ComprehendGenerate structures that relate stimulus to tasks and can be used to inform behavior TaskPerform task maintenance IntendChoose an action, create prediction DecodeDecompose action into motor commands MotorExecute motor commands

Newell’s Abstract Functional Operations (Newell 1990) 9 …but how these actually work was not clear. PerceiveWhat information is generated? EncodeWhat information is generated? AttendWhat information is required? ComprehendWhat information is required and generated? TaskWhat information is required? IntendWhat information is required?

NAFO and Appraisal (Marinier & Laird 2006) 10 Generated ByRequired By Suddenness Perceive Attend Unpredictability Encode Intrinsic pleasantness Goal relevance Causal agent Comprehend Comprehend, Task, Intend Causal motive Outcome probability Discrepancy from expectation Goal/need conduciveness Control Power

Integrate emotion with a complete agent framework Computationally distinguish emotion, mood and feeling Weight feeling’s importance by computing its intensity Use feeling as intrinsic reward signal to drive reinforcement learning Discussion & Conclusion Outline 11 Situation Goals Appraisal Emotion, Mood, Feeling Coping

Body Symbolic Long-Term Memories Procedural Short-Term Memory Situation, Goals Decision Procedure Chunking Reinforcement Learning Semantic Learning Episodic Learning Perception Action Visual Imagery Feeling Generation Extending Soar with Emotion (Marinier & Laird 2007) 12 Soar is a cognitive architecture A cognitive architecture is a set of task-independent mechanisms that interact to give rise to behavior Cognitive architectures are general agent frameworks

Feeling Generation Reinforcement Learning Emotion.5,.7,0,-.4,.3,… Extending Soar with Emotion (Marinier & Laird 2007) 13 Body Decision Procedure Perception Action Appraisals Feelings Short-Term Memory Situation, Goals Mood.7,-.2,.8,.3,.6,… Feelings Knowledge Architecture Symbolic Long-Term Memories Procedural Chunking Semantic Learning Episodic Learning +/- Intensity Feeling.9,.6,.5,-.1,.8,… Visual Imagery

Computing Feeling from Emotion and Mood (Marinier & Laird 2007) 14 Assumption: Appraisal dimensions are independent Limited Range: Inputs and outputs are in [0,1] or [-1,1] Distinguishability: Very different inputs should lead to very different outputs Non-linear: Linearity would violate limited range and distinguishability

Example 15 EmotionMoodFeeling Suddenness [0,1]0.235 Unpredictability [0,1] Intrinsic-pleasantness [-1,1] Goal-relevance [0,1] Causal-agent (self) [0,1]000 Causal-agent (other) [0,1]000 Causal-agent (nature) [0,1] Causal-motive (intentional) [0,1]000 Causal-motive (chance) [0,1] Causal-motive (negligence) [0,1]000 Outcome-probability [0,1] Discrepancy [0,1] Conduciveness [-1,1] Control [-1,1] Power [-1,1] Labelela-joyanx-worela-joy

Example 16

Maze Task Start Goal 17

Feeling Dynamics Results 18 very easy

Computing Feeling Intensity (Marinier & Laird 2007) 19 Motivation: Intensity gives a summary of how important (i.e., how good or bad) the situation is Limited range: Should map onto [0,1] No dominant appraisal: No single value should drown out all the others Can’t just multiply values, because if any are 0, then intensity is 0 Realization principle: Expected events should be less intense than unexpected events

Example 20 EmotionMoodFeeling Suddenness [0,1]0.235 Unpredictability [0,1] Intrinsic-pleasantness [-1,1] Goal-relevance [0,1] Causal-agent (self) [0,1]000 Causal-agent (other) [0,1]000 Causal-agent (nature) [0,1] Causal-motive (intentional) [0,1]000 Causal-motive (chance) [0,1] Causal-motive (negligence) [0,1]000 Outcome-probability [0,1] Discrepancy [0,1] Conduciveness [-1,1] Control [-1,1] Power [-1,1] Labelela-joyanx-worela-joy Intensity.127

Example 21

Outline 22 Integrate emotion with a complete agent framework Computationally distinguish emotion, mood and feeling Weight feeling’s importance by computing its intensity Use feeling as intrinsic reward signal to drive reinforcement learning Discussion & Conclusion

Intrinsically Motivated Reinforcement Learning (Sutton & Barto 1998; Singh et al. 2004) 23 Environment Critic Agent ActionsStatesRewards External Environment Internal Environment Agent Critic Actions StatesRewardsDecisions Sensations Appraisal Process +/-Feeling Intensity “Organism”

Learning Task Start Goal 24

Learning Results 25

Discussion & Conclusion 26 Discussion Agent learns fast Gets frequent reward signals Mood accelerates learning Provides reward during those steps in which the agent has no emotion Conclusion Developed an initial computational model of emotion Integrated model with complete agent framework Demonstrated some functional advantages of integration