Sarit Kraus BarIlan University saritcsbiuacil BuyerSeller Interaction Buyers and sellers across geographical and ethnic borders Electronic commerce Crowdsourcing Automated travel agents ID: 322904
Download Presentation The PPT/PDF document "Automated Agents that Interact Proficien..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Automated Agents that Interact Proficiently with People
Sarit KrausBar-Ilan University
sarit@cs.biu.ac.ilSlide2Slide3
Buyer-Seller Interaction
Buyers and sellers across geographical and ethnic bordersElectronic commerce Crowd-sourcing Automated travel agents
BargainingSlide4
Culture Sensitive Agents
The development of standardized agents to be used in the collection of data for studies on culture and negotiation
BargainingSlide5
Automated Mediators for Resolving Conflicts
BargainingSlide6
Medical Applications: Rehabilitation & Care
Reinforcement for rehabilitation in an inpatient rehabilitation unitPersonalized automated speech therapist
6
6
Sheba
Hospital
PersuasionSlide7
Medical Applications:
Preventing Unhealthy Behaviors
PersuasionSlide8
Sustainability: Reducing Fuel Consumption
PersuasionSlide9
Advice Provision for Decision Making
Discussion AgentSlide10
Training People
Virtual suspect to
train investigators
Training people in
n
egotiations
(employer-employee)Slide11
Why not Equilibrium Agents?
Nash equilibrium: stable strategies; no agent has an incentive to deviateResults from the social sciences suggest people do not follow equilibrium strategies:Equilibrium based agents played against people failed.
People rarely design agents to follow equilibrium strategies. Slide12
People
Often Follow Suboptimal Decision StrategiesIrrationalities attributed tosensitivity to contextlack of knowledge of own preferences
the effects of complexity
the interplay between emotion and cognition
the problem of self control Slide13
Why not Only Behavioral Science Models?
There are several models that describe human decision makingMost models specify general criteria that are context sensitive but usually do not provide specific parameters or mathematical definitionsSlide14
Why not Only Machine Learning?
Machine learning builds models based on dataIt is difficult to collect human data Collecting data on specific user is very time consuming.Human data is noisy
“Curse” of dimensionalitySlide15
Methodology
Human Prediction Model
Take action
machine learning
Game Theory Optimization
methods
Data
(from specific culture)
Human behavior models
Human specific dataSlide16
Predicting Human Decisions
ActionsDrivers choicesNegotiators’ reliabilityInvestors/investeesVoters
Text
Pirate game
Facial expressions Slide17
What is she going to do? Stay or LeaveSlide18
Successes?
Security in LAXSlide19
Slide20
Human behavior model
Human behavior models
Take action
M
achine
learning
Game Theory Optimization
methods
Data
(from specific culture)
Challenging:
Experimenting
with people is very difficult !!!
Working with people from other disciplines is challenging.
Agents interacting proficiently with people is important
Fun
Human specific data
Challenging:
How
to integrate machine learning and behavioral
models?
How to use in agent’s strategy?
sarit@cs.biu.sc.il