Full of feeling Figuring: Specialists With Feeling.

Uploaded on:
Plan. IntroductionHighlighted ProjectsAffective Cognitive Learning
Slide 1

Full of feeling Computing: Agents With Emotion Victor C. Hung University of Central Florida – Orlando, FL EEL6938: Special Topics in Autonomous Agents March 29, 2007

Slide 2

Agenda Introduction Highlighted Projects Affective Cognitive Learning & Decision Making Questions

Slide 3

Introduction Affective Computing identifies with, emerges from, or intentionally impacts feeling or other full of feeling wonders Engineering, software engineering with brain science, intellectual science, neuroscience, humanism, instruction, psychophysiology, morals … Emotion is essential to human experience Cognition Perception Learning Communication Rational basic leadership

Slide 4

Introduction Technologists have to a great extent overlooked feeling Affect has been misjudged Hard to gauge MIT Media Lab: Affective Computing http://affect.media.mit.edu Develop new innovations and hypotheses Understanding effect and its part in human experience Restore an appropriate harmony amongst feeling and cognizance in the configuration of advances for tending to human needs

Slide 5

Introduction Issues in full of feeling registering Communication of emotional subjective states to machines Techniques to survey dissatisfaction, anxiety, and disposition by implication Make PCs can be all the more sincerely wise Personal advances for enhancing mindfulness of full of feeling states Emotion\'s impacts individual wellbeing Ethics

Slide 6

Highlighted Projects Affective-Cognitive Framework for Machine Learning and Decision-Making Emotion\'s part in learning and basic leadership Digital Story Explication as it Relates to Emotional Needs and Learning Emotional association in tyke learning ESP - The Emotional-Social Intelligence Prosthesis Aid for the inwardly impeded

Slide 7

Highlighted Projects Fostering Affect Awareness and Regulation in Learning Combat disappointment amid the learning procedure Machine Learning and Pattern Recognition with Multiple Modalities Emotional sensor information combination Ripley: A Conversational Robot Human-robot collaboration stage through dialect and visual observation modalities

Slide 8

Affective-Cognitive Learning & Decision Making (2006) Ahn and Picard\'s " Affective-Cognitive Learning and Decision Making: The Role of Emotions ", The eighteenth European Meeting on Cybernetics and Systems Research Framework for learning and basic leadership Inspired by neural premise of inspirations and the part of feelings in human conduct Affective inclinations Loss abhorrence Effect of temperament on basic leadership

Slide 9

Affective-Cognitive Learning & Decision Making Affective predispositions Two-outfitted scoundrel

Slide 10

Affective-Cognitive Learning & Decision Making Loss revultion Prefer maintaining a strategic distance from misfortunes than getting picks up

Slide 11

Affective-Cognitive Learning & Decision Making Effect of mind-set on basic leadership ANGER Optimism about the future HAPPINESS Optimism about the present Pessimism about the future FEAR Pessimism about the present SADNESS

Slide 12

Affective-Cognitive Learning & Decision Making A motivational quality (prize)- based learning hypothesis: Extrinsic worth from the psychological (deliberative and systematic) frameworks Intrinsic worth from different emotional frameworks, for example, Seeking (Wanting), Fear, Rage, and different circuits Probabilistic models Cognition (subjective state move) Multiple influence circuits (Seeking, Joy, Anger, Fear, ...) Decision making model Previous learning can be joined for expecting the outcomes of choices (or figuring the psychological worth)

Slide 13

Affective-Cognitive Learning & Decision Making

Slide 14

Affective-Cognitive Learning & Decision Making The Decision-Making Model Cognitive state (c) Affective state (a) Decision (d)

Slide 15

Affective-Cognitive Learning & Decision Making Affective looking for worth = Valence = chose by the mean of the separated qualities for the prize specimens Arousal = instability of the prize example circulation (demonstrated as standard deviation) Complete basic leadership expression: Non-influence operator has just the psychological segment

Slide 16

Affective-Cognitive Learning & Decision Making Affective operator versus Non-influence operator

Slide 17

Affective-Cognitive Learning & Decision Making Influence of an exception on the subjective qualities and the valence values

Slide 18

Affective-Cognitive Learning & Decision Making Affective part less delicate to anomalies than psychological segment Affective Cooling: Agreement between two segments More prone to take after the choice by the intellectual segment (Exploitation) Value of the instigated backwards temperature parameter expands Humans utilizing perception as a part of basic leadership Affective Heating: Conflict between two segments Less liable to take after the choice by the intellectual segment (Exploration) Value of the prompted opposite temperature parameter diminishes Humans relying upon feeling in basic leadership

Slide 19

Affective-Cognitive Learning & Decision Making 10-furnished desperado assignments

Slide 20

Affective-Cognitive Learning & Decision Making Too much or too little influence weakens learning Excessive adapts speedier, yet not useful for long haul Insufficient better for long haul, yet moderate

Slide 21

Affective-Cognitive Learning & Decision Making Results and Conclusions Framework improvements Model other influence circuits Incidental impacts on basic leadership Use of earlier information for expecting psychological results ・ Affective inclination Helps consequently direct investigation and abuse Speed up learning without yielding choice quality This structure may copy very much contemplated human conduct Risk revultion Effects of mind-set on basic leadership Self-control

Slide 22


View more...