Evolving Agents for the Hanabi 2018 CIG Competition

Canaan, Rodrigo; Shen, Haotian; Torrado, Ruben Rodriguez; Togelius, Julian; Nealen, Andy; Menzel, Stefan

doi:10.1109/cig.2018.8490449

Cited by 17 publications

(23 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The NYUGameLab agent shows a more significant drop in performance in larger games. This is consistent with the results presented in their competition paper [19]. The results from the mixed track show a much smaller difference in agent performance, table III shows that although MonteCarloOppNN performed slightly better, there is less than a point between the agents.…”

Section: Competition Results and Analysissupporting

confidence: 90%

See 1 more Smart Citation

The 2018 Hanabi competition

Walton-Rivers

Williams

Bartle

2019

2019 IEEE Conference on Games (CoG)

View full text Add to dashboard Cite

This paper outlines the Hanabi competition, first run at CIG 2018, and returning for COG 2019. Hanabi presents a useful domain for game agents which must function in a cooperative environment. The paper presents the results of the two tracks which formed the 2018 competition and introduces the learning track, a new track for 2019 which allows the agents to collect statistics across multiple games.

show abstract

Section: Competition Results and Analysissupporting

confidence: 90%

“…NYUGameLab agent, created by Canaan, Shen, Torrado, et al [19]. The agent is a combination of the rule-based agent framework supplied with the framework combined with a genetic algorithm.…”

Section: B Nyugamelabmentioning

confidence: 99%

The 2018 Hanabi competition

Walton-Rivers

Williams

Bartle

2019

2019 IEEE Conference on Games (CoG)

View full text Add to dashboard Cite

show abstract

“…There were two tracks, called "Mirror" and "Mixed" which broadly match the two categories we propose in Section 3. Similarly to van den Bergh, the second-place player used a genetic algorithm to evolve a sequence of rules from a fixed rule set [53]. This agent achieved an average score of 17.52 points in the Mirror competition, while the first place agent, "Monte Carlo NN", achieved a score of 20.57.…”

Section: Prior Work On Hanabi Aimentioning

confidence: 99%

The Hanabi challenge: A new frontier for AI research

Bard¹,

Foerster

Chandar

et al. 2020

Artificial Intelligence

157

128

View full text Add to dashboard Cite

From the early days of computing, games have been important testbeds for studying how well machines can do sophisticated decision making. In recent years, machine learning has made dramatic advances with artificial agents reaching superhuman performance in challenge domains like Go, Atari, and some variants of poker. As with their predecessors of chess, checkers, and backgammon, these game domains have driven research by providing sophisticated yet well-defined challenges for artificial intelligence practitioners. We continue this tradition by proposing the game of Hanabi as a new challenge domain with novel problems that arise from its combination of purely cooperative gameplay with two to five players and imperfect information. In particular, we argue that Hanabi elevates reasoning about the beliefs and intentions of other agents to the foreground. We believe developing novel techniques for such theory of mind reasoning will not only be crucial for success in Hanabi, but also in broader collaborative efforts, especially those with human partners. To facilitate future research, we introduce the open-source Hanabi Learning Environment, propose an experimental framework for the research community to evaluate algorithmic advances, and assess the performance of current state-of-the-art techniques. 6 One such equilibrium occurs when players do not intentionally communicate information to other players, and ignore what other players tell them (historically called a pooling equilibrium in pure signalling games [15], or a babbling equilibrium in later work using cheap talk [16]). In this case, there is no incentive for a player to start communicating because they will be ignored, and there is no incentive to pay attention to other players because they are not communicating.7 In pure signalling games where actions are purely communicative, policies are often referred to as communication protocols. Though Hanabi is not such a pure signalling game, when we want to emphasize the communication properties of an agent's policy we will still refer to its communication protocol. 8 We use the word convention to refer to the parts of a communication protocol or policy that interrelate. Technically, these can be thought of as constraints on the policy to enact the convention.

show abstract

“…Our 2018 competition entry, which took second place, is described in [17]. We implemented an evolutionary algorithm to make rule-based agents by searching for a well-performing sequence of rules both for self-play and mixed play, using the same pool as [13] for mixed play.…”

Section: B Hanabi-playing Agents and The Competitionmentioning

confidence: 99%

“…We use a similar representation of individuals as the one we used in [17]. Each individual is represented by a chromosome defined by a sequence of 15 integers, each integer representing one of 135 possible rules.…”

Section: B Representation and Operatorsmentioning

confidence: 99%

Diverse Agents for Ad-Hoc Cooperation in Hanabi

Canaan

Togelius

Nealen

et al. 2019

2019 IEEE Conference on Games (CoG)

Self Cite

View full text Add to dashboard Cite

In complex scenarios where a model of other actors is necessary to predict and interpret their actions, it is often desirable that the model works well with a wide variety of previously unknown actors. Hanabi is a card game that brings the problem of modeling other players to the forefront, but there is no agreement on how to best generate a pool of agents to use as partners in ad-hoc cooperation evaluation. This paper proposes Quality Diversity algorithms as a promising class of algorithms to generate populations for this purpose and shows an initial implementation of an agent generator based on this idea. We also discuss what metrics can be used to compare such generators, and how the proposed generator could be leveraged to help build adaptive agents for the game.

show abstract

Evolving Agents for the Hanabi 2018 CIG Competition

Cited by 17 publications

References 15 publications

The 2018 Hanabi competition

The 2018 Hanabi competition

The Hanabi challenge: A new frontier for AI research

Diverse Agents for Ad-Hoc Cooperation in Hanabi

Contact Info

Product

Resources

About