Our great sponsors
-
dartmcts
A Monte Carlo Tree Search library written in Dart used to power opponents in the Trickster's Table app
-
InfluxDB
Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.
I've been working on the Trickster's Table app for the past six months or so. It's an app written in Dart/Flutter that lets you can play modern trick taking games against AIs trained using reinforcement learning.
The app currently contains implementations of Boast or Nothing, Yokai Septet (2-player), Potato Man, and (on beta channels) Magic Trick. Here's a Geeklist showing other games I'm planning to add: https://boardgamegeek.com/geeklist/226363/tricksters-table-a...
Certain parts of it are open source, including the most recently added game engine for the game Magic Trick by Chris Wray: https://github.com/dbravender/magictrick and the Monte Carlo Tree Search / Neural Network library it uses: https://github.com/dbravender/dartmcts
Currently, for training the AIs I use this PR against SIMPLE so I don't have to write each game in both Dart and Python: https://github.com/davidADSP/SIMPLE/pull/34
I've got a long list of things to fix and try when it comes to training the AIs (including trying integrating with other frameworks to see how well they perform). Is anyone interested in helping to train AIs or implementing new games? I can check with the designers of some of the existing games or future games to see if they are okay with me sharing the source code for their games.
iOS: https://apps.apple.com/us/app/tricksters-table/id1668506875 TestFlight to get the latest games (including Magic Trick): https://testflight.apple.com/join/uuFXY9mw
I've been working on the Trickster's Table app for the past six months or so. It's an app written in Dart/Flutter that lets you can play modern trick taking games against AIs trained using reinforcement learning.
The app currently contains implementations of Boast or Nothing, Yokai Septet (2-player), Potato Man, and (on beta channels) Magic Trick. Here's a Geeklist showing other games I'm planning to add: https://boardgamegeek.com/geeklist/226363/tricksters-table-a...
Certain parts of it are open source, including the most recently added game engine for the game Magic Trick by Chris Wray: https://github.com/dbravender/magictrick and the Monte Carlo Tree Search / Neural Network library it uses: https://github.com/dbravender/dartmcts
Currently, for training the AIs I use this PR against SIMPLE so I don't have to write each game in both Dart and Python: https://github.com/davidADSP/SIMPLE/pull/34
I've got a long list of things to fix and try when it comes to training the AIs (including trying integrating with other frameworks to see how well they perform). Is anyone interested in helping to train AIs or implementing new games? I can check with the designers of some of the existing games or future games to see if they are okay with me sharing the source code for their games.
iOS: https://apps.apple.com/us/app/tricksters-table/id1668506875 TestFlight to get the latest games (including Magic Trick): https://testflight.apple.com/join/uuFXY9mw
I've been working on the Trickster's Table app for the past six months or so. It's an app written in Dart/Flutter that lets you can play modern trick taking games against AIs trained using reinforcement learning.
The app currently contains implementations of Boast or Nothing, Yokai Septet (2-player), Potato Man, and (on beta channels) Magic Trick. Here's a Geeklist showing other games I'm planning to add: https://boardgamegeek.com/geeklist/226363/tricksters-table-a...
Certain parts of it are open source, including the most recently added game engine for the game Magic Trick by Chris Wray: https://github.com/dbravender/magictrick and the Monte Carlo Tree Search / Neural Network library it uses: https://github.com/dbravender/dartmcts
Currently, for training the AIs I use this PR against SIMPLE so I don't have to write each game in both Dart and Python: https://github.com/davidADSP/SIMPLE/pull/34
I've got a long list of things to fix and try when it comes to training the AIs (including trying integrating with other frameworks to see how well they perform). Is anyone interested in helping to train AIs or implementing new games? I can check with the designers of some of the existing games or future games to see if they are okay with me sharing the source code for their games.
iOS: https://apps.apple.com/us/app/tricksters-table/id1668506875 TestFlight to get the latest games (including Magic Trick): https://testflight.apple.com/join/uuFXY9mw