Computers that can Negotiate ERCIM Cor Baayen Award Tim Baarslag - - PowerPoint PPT Presentation
Computers that can Negotiate ERCIM Cor Baayen Award Tim Baarslag Researcher in Centrum Wiskunde & Informatica (CWI), Research institute for Mathematics and Computer Science in the Netherlands Negotiation Negotiation is everywhere
Tim Baarslag
Researcher in Centrum Wiskunde & Informatica (CWI), Research institute for Mathematics and Computer Science in the Netherlands
ERCIM Cor Baayen Award
– Leaving money on the table, bounded rationality – Biases & emotions, time & costs
Research line I
3
and opponent
𝑪
Good for me Good for opponent
5
Contract 𝑩 Contract 𝑪
Fte: 1.0 Salary: $3500 Car: no
𝑩
Fte: 0.8 Salary: $3000 Car: yes
dominanto utcomes
Good for me Good for opponent
6
bidding accepting learning
Baarslag et al., Evaluating Practical Negotiating Agents: Results and Analysis of the 2011 International Competition, Artificial Intelligence, 2013.
Research line II
8
many digital services
– Take it or leave it approach – One size fits all – Opaque business models
incomplete preferences
User utility Opponent utility
10
dominanto utcomes
State-of-the-art
no uncertainty
11
User utility Opponent utility
State-of-the-art
no uncertainty
uncertainty
12
User utility Opponent utility
– Elicitation on-the-fly: which queries to ask? – What is (costly) user information worth?
Unexplored State-of-the-art
no uncertainty full uncertainty
uncertainty
Reduce uncertainty with costly queries
published data
more meaningful consent
Baarslag et al., Negotiation As an Interaction Mechanism for Deciding App Permissions, CHI Late Breaking Work, 2016.
management
Baarslag et al. How artificial intelligence could negotiate better deals for humans. Science, 2017. Baarslag et al. How would a machine conduct our salary negotiations? Wired, 2017. Lewis et al. Deal or No Deal? End-to-End Learning for Negotiation Dialogues. Facebook AI, 2017.