webDiplomacy VS nle

Thanks to sweetandcool for creating WPL_1.1 for me on webDiplomacy.

I am organising Gunboat Leagues on Backstabbr, Playdiplomacy and webDiplomacy.

I really wish webdiplomacy.net would clean up their website to be more professional looking like this.

Playdiplomacy runs mentor games every once in a while. I've run them there before - I don't have time at the moment. webDiplomacy had a similar type of game, too - don't know if they still do. He might want to try these. To be fair, though, with the ODC in the 3rd and 4th rounds, Nexus and Olympus tournaments hitting the semi-finals, finding someone to run this type of game might be difficult.

In late November, Meta announced Cicero, an AI agent that can beat humans at the strategy board game Diplomacy in online games played on webDiplomacy.net. That's a major achievement because Diplomacy is a largely social game that requires extensive persuasion, cooperation, and negotiation with other players to win. Basically, Meta developed a bot that could fool humans into thinking they were playing with another human.

Since this is a research effort, we don't have plans to host CICERO for public availability. However, we have open-sourced both the model files and code, which means you could host CICERO yourself on a private instance of webDiplomacy.net (also open sourced here). More details can be found here.

I was casually interested to know if one could play at this game, and I found out that you can go single player vs. bots here:

https://webdiplomacy.net/

You can play Diplomacy via web at https://webdiplomacy.net/

Have you tried Diplomacy? https://en.wikipedia.org/wiki/Diplomacy_(game) JFK’s fave game, apparently. Have done several online games with friends — took 2-3 months each with 24-48 hour turns. Wild stuff, got very intense https://webdiplomacy.net/

precisely, I really hope someone does Nethack next and leverages the learning environment that's already customized for it.

You should look at The Nethack Learning Environment.

There's quite a few open-source Reinforcement Learning challenges that you can explore with modest amounts of compute in order to build some experience training RL models, for example the Nethack Learning Environment, Atari, Minigrid, etc. For me personally, I had only worked in NLP / dialogue for years but got into RL by implementing Random Network Distillation models for NetHack. It's a fun area that definitely has its own unique challenges vs other domains. -AM

The NetHack Learning Environment\ (2 comments)