• Arun Selvi K

Games were relevant to research to investigate how far computers would handle complex decision-making from the early days of computing. Machine learning has made considerable strides over recent years with automated staff that attain incredible success in difficult fields such as Go, Atari and other poker variants. Like their predecessors of chess, scrutinizers and backgammon, some areas of game science have presented artificial intelligence practitioners with complex yet well-established obstacles. The researchers continue this practice by introducing the Hanabi game as a new difficulty field with new difficulties resulting from the mixture of strictly cooperative gaming with 2 to 5 players and incomplete details. They contend in particular that Hanabi puts thinking on certain agents' beliefs and motives in the forefront. This paper assumes that creating new strategies for the philosophy of learning in Hanabi, and in particular those with human partners would be crucial to progress. To facilitate further research, this paper incorporates the Hanabi Learning Environment open-source, provides a research community with an investigational foundation for assessing algorithmic developments and determines the efficiency of modern technologies.

Artificial Intelligence, Hanabi Challenge, Imperfect Information, Multi-Agent Learning, Reinforcement Learning, Game Theory of Mind.

"Review on Hanabi Challenge for Artificial Intelligence Research ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 2, page no.424-429, February-2019, Available :http://www.jetir.org/papers/JETIREV06067.pdf

"Review on Hanabi Challenge for Artificial Intelligence Research ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 2, page no. pp424-429, February-2019, Available at : http://www.jetir.org/papers/JETIREV06067.pdf