[agents] CFP: The Automated Negotiating Agents Competition 2019 - Agent Negotiation under Preference Uncertainty

[Tim Baarslag]

=================================================================================
Call for participation: The Automated Negotiating Agents Competition 
2019 - Agent Negotiation under Preference Uncertainty

Challenge: Representing users in a negotiation - developing an agent 
that can negotiate under preference uncertainty.

Submission deadline: 20 May, 2019

Event: IJCAI 2019, August 10-16 2019, in Macao, China

Website: http://web.tuat.ac.jp/~katfuji/ANAC2019/#genius
=================================================================================


====== Challenge ======
The challenge for ANAC 2019 is to design an agent that can negotiate 
under preference uncertainty. The idea is that when a negotiating agent 
represents a user in a negotiation, it cannot know exactly what the user 
wants due to practical limits on the preference elicitation process.

In previous years of ANAC, and in most literature on automated 
negotiation, the utility function of the agent is presumed known. 
Instead, this year, the preferences of the agent will be given in the 
form of a ranking of a limited number of possible agreements.


====== Event ======
The competition takes place during IJCAI 2019, August 10-16 2019, in 
Macao, China. The prize money for the competition is at least $5000 in 
total, and several student travel grants will be made available to 
attend the competition.


====== Entrants ======
Entrants to the competition have to develop and submit an autonomous 
negotiating agent that runs on Genius. Genius is a Java-based 
negotiation platform in which you can develop general negotiating agents 
as well as create negotiation domains and preference profiles. The 
platform allows you to simulate negotiation sessions and run 
tournaments. Example agents are available for the ANAC 2019 setting. 
More details can be found by following this link: 
http://ii.tudelft.nl/genius/

Performance of the agents will be evaluated in a tournament between all 
participants, where each agent is matched with other submitted agents. 
All submitted agents will negotiate in a number of negotiation 
scenarios, with varying levels of preference uncertainty.


====== Negotiating with partial preferences ======
This year, rather than having access to a utility function, the 
preferences of the agent will be given in the form of a ranking of 
outcomes (i.e. w1 <= ... <= wd, where each wi is a possible agreement).

The rankings are randomly generated from existing negotiation scenarios 
in which full utility information is available from a standard linearly 
additive utility function u. The number of rankings d that the agent 
receives represents the amount of preference uncertainty of the agent. 
The agent negotiates with these ordinal preferences until a certain 
outcome w* is reached. The score the agent will receive for this 
agreement will be based on the original utility u(w*) that is associated 
with it. That is, the agent receives ordinal information only, but it 
will be evaluated based on the underlying cardinal ground-truth.

The challenge for the agents is to estimate the best fitting linear 
utility function from the ranking, using techniques from e.g. machine 
learning, regression techniques, trade-offs, and linear programming.

For details, please refer to the Genius manual and the frequently asked 
questions at: http://tinyurl.com/ANAC2019GeniusFAQ


====== Rules of Encounter ======
Negotiations are bilateral encounters following the alternating offers 
protocol. Offers are exchanged using a round-based protocol, with a 
maximum of 1000 rounds. Agents do not have any prior knowledge about the 
preferences and strategy of the opponent. Agents are reset after each 
encounter; that is, agents may negotiate repeatedly on the same domain 
or with the same opponents, but they cannot learn from their previous 
negotiations. When no agreement is struck, both agents receive their 
reservation value, which is privately known utility value between 0 and 
1. The reservation value can be different for both players. Note that 
this means a break-off can be preferable to an agreement for one (or 
both) of the players. This implies that it is risky to wait until the 
deadline to reach an agreement. This year features no discount factor.

The alternating offers protocol specifies that both negotiators take 
turns making offers and is implemented as a special bilateral case of 
the multilateral Stacked Alternating Offers Protocol. One agent starts 
the negotiation with an opening bid, after which the other party can 
take the following actions:

     - Make a counter offer (thus rejecting and overriding the previous 
offer);
     - Accept the offer;
     - Walk away (i.e. ending the negotiation without any agreement and 
receiving the reservation value).

This process ends until either an (dis)agreement is reached or 1000 
rounds of exchanges have been made. If no agreement has been reached 
before this time, the negotiation will end in a break-off. In order to 
make running the tournament feasible, agents are expected to generate 
their offers in a reasonable amount of time. Agents that take too long 
will have a time-out imposed on the negotiation after approximately one 
minute of real time (the exact moment is not given to avoid strategic 
considerations around the real-time clock).

Agents will be disqualified for violating the spirit of fair play (e.g. 
hacking the API, starting threads, attempting to access other party’s 
preference profile). The board of the ANAC 2019 competition will be the 
judge on these matters (for more information, see 
http://ii.tudelft.nl/anac/).


====== Evaluation ======
The winners will be determined in two separate categories: the average 
individual utilities gained by each agent, and the average social 
welfare (i.e., average product of utilities of both agents). The teams 
of the top performing agents will be notified, and the final results and 
awards will be announced at IJCAI 2019. It is expected that teams that 
make it through to the finals will have a representative attending the 
conference.


====== Submission (Deadline: 20 May, 2019) ======
The competition rules allow multiple entries from a single institution, 
but require each agent to be developed independently. Participants 
submit their agent source code and class files (in a .zip or .jar 
package) through the following link: https://tinyurl.com/GENIUSANAC2019


====== Academic report ======
Each participant has the option to prepare a 2-4 page report describing 
the design of heir agent according to academic standards. The best teams 
that submit a report will be given the opportunity to give a brief 
presentation describing their agent at IJCAI. Furthermore, proceedings 
of the competition are planned to be published by Springer in the 
Studies in Computational Intelligence (SCI) series.

The report will be evaluated by the organizers of the league. For 
eligibility, the strategy design should provide a contribution to the 
negotiation community. The report is recommended to address the 
following aspects:

     - Bidding Strategy: how the agent generates bid at its each turn;
     - Acceptance Strategy: how the agent decides to accept or reject a 
given bid;
     - Opponent Modelling: how the agent models the opponent (e.g. the 
opponent’s strategy, preferences etc.);
     - Preference uncertainty: how the agent deals with preference 
uncertainty, explaining which heuristics and/or machine learning methods 
are employed;
     - Evaluation: an evaluation of the agent (either against itself or 
in a small-scale tournament setting).


====== Important Dates ======
Submission deadline: May 20, 2019
Notification to finalists: June 15, 2019 (tentative)
Event: August 10-16, 2019


====== Questions and Answers ======
Website: http://web.tuat.ac.jp/~katfuji/ANAC2019/#genius
Feel free to ask your questions in the FAQ, see the following link: 
http://tinyurl.com/ANAC2019GeniusFAQ

Send your questions to:
     T.Baarslag at cwi.nl (main contact)
     reyhan.aydogan at ozyegin.edu.tr
     katfuji at cc.tuat.ac.jp