ΑΙhub.org
 

#IJCAI2019 mini-interviews – Claus Aranha from University of Tsukuba

by
14 August 2019



share this:

Meet Claus Aranha, Assistant Professor at the University of Tsukuba, Center of Artificial Intelligence Research (C-AIR).

What are you presenting at IJCAI?
I am organizing the werewolf track ANAC Competition.

 

Can you tell me more about ANAC?
ANAC stands for Automated Agent Negotiation Competition<sup>1</sup>. It is a competition where AI Agents negotiate with each other and/or humans. We have 6 leagues this year – Supply Chain Management, Werewolf (a social party game), Diplomacy (a board game), Human-Agent League, Agent-Agent League: Agent Negotiation with Partial Preferences and the GENIUS league.

 

What is the real world impact of agents negotiating with each other?
Take for example the Supply Chain Management (SCM) league. Right now in the SCM industry many stakeholders – from people who sell the raw material, to factories which produce goods, to shops who sell them are involved. Each one has their preferences on what they wish to get out of the deal and what they are ready to compromise on. They have to coordinate (i.e. negotiate) with each other all the time for timelines, prices, quantities, etc. It is a cumbersome and cognitively heavy job! 

Imagine a group of AI agents representing each stakeholder does this for you. Wouldn’t things become easier?

 

You said something about the werewolf track. Could you say more?
It is a social party game where agents are supposed to find who the werewolf is<sup>2</sup> in a setting where agents lie to each other and/or hide the truth about themselves. How can an agent deal with other agents who behave this way while the agent itself is also deceiving other agents — It is a hard challenge!

We had 90 teams this year out of which 70 sent an agent to the competition. We have 15 finalists and the top three will be discussed tomorrow (August 15)! Keep an eye out 🙂

 

How can I get involved?
Check out the AI Wolf project page3. You can also get some sample code here – https://github.com/caranha/AIWolfCompo

 

1http://web.tuat.ac.jp/~katfuji/ANAC2019/

2https://en.wikipedia.org/wiki/Mafia_(party_game)

3http://aiwolf.org/en/




Rahul Divekar is a PhD Candidate at the Department of Computer Science at Rensselaer Polytechnic Institute.
Rahul Divekar is a PhD Candidate at the Department of Computer Science at Rensselaer Polytechnic Institute.




            AIhub is supported by:


Related posts :



Keeping learning-based control safe by regulating distributional shift

We propose a new framework to reason about the safety of a learning-based controller with respect to its training distribution.
30 September 2022, by

Bipedal robot achieves Guinness World Record in 100 metres

Cassie the robot, developed at Oregon State University, records the fastest 100 metres by a bipedal robot.
29 September 2022, by

#IJCAI2022 distinguished paper – Plurality veto: A simple voting rule achieving optimal metric distortion

How can we create a voting system that best represents the preferences of the voters?
28 September 2022, by

AIhub monthly digest: September 2022 – environmental conservation, retrosynthesis, and RoboCup

Welcome to our monthly digest, where you can catch up with AI research, events and news from the month past.
27 September 2022, by

The Machine Ethics Podcast: Rights, trust and ethical choice with Ricardo Baeza-Yates

Host Ben Byford chats to Ricardo Baeza-Yates about responsible AI, the importance of AI governance, questioning people's intent to create AGI, and more.
26 September 2022, by

Recurrent model-free RL can be a strong baseline for many POMDPs

Considering an approach for dealing with realistic problems with noise and incomplete information.
23 September 2022, by





©2021 - Association for the Understanding of Artificial Intelligence


 












©2021 - ROBOTS Association