News for hackers. Stuff that matters.

A comment on The Hanabi Challenge: A New Frontier for AI Research.

Re:The Hanabi Challenge: A New Frontier for AI Research by knicholes on Tuesday February 05, 2019 @02:06PM (#19088601)
I wonder why they created their own OpenAI-like interface instead of inheriting from OpenAI's environment?
[ Parent | on HN ]
Re:The Hanabi Challenge: A New Frontier for AI Research by jakobnicolaus on Tuesday February 05, 2019 @03:00PM (#19089178)
Hanabi is a multi-agent problem. Unfortunately gym doesn't natively support multi-agent action and state spaces.
[ Parent | on HN ]
Re:The Hanabi Challenge: A New Frontier for AI Research by soraki_soladead on Tuesday February 05, 2019 @02:38PM (#19088933)
Two reasons come to mind:

DeepMind and OpenAI are indirectly in competition with each other.

The Gym interface is okay but the implementation is poor and does not readily support multiple agents.

[ Parent | on HN ]