irwin
maia-chess
Our great sponsors
irwin | maia-chess | |
---|---|---|
57 | 97 | |
486 | 875 | |
- | 3.1% | |
0.0 | 1.6 | |
over 1 year ago | 8 months ago | |
Python | Python | |
GNU Affero General Public License v3.0 | GNU General Public License v3.0 only |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
irwin
-
How common is false banning in chess?
For Lichess, you can get some sort of idea from reading the code. If I remember correctly, there's some threshold for the site suspecting you of cheating based on a variety of simple metrics (accuracy, blurring, etc) and then it gets sent to machine learning tools to analyze (here is one of them, and here's another).
-
Spotting a cheater: Stats analysis
Yeah Irwin is the old lichess model - here (https://github.com/clarkerubber/irwin) as well as the more maintained fork (https://github.com/lakinwecker/irwin/commits/master)
- A question for technologists: can we start an open-source cheat-detection engine that becomes the gold standard of cheat detection engines?
- Can we see the Lichess cheat detection stuff?
-
Machine Learning for detecting anomalies in chess
Isnt't lichess' cheat detection an ML based system?
-
Governance bodies as private companies: a wrinkle in the Carlsen-Niemann drama
It's probably more of a security by obscurity thing - withholding technical details in order to protect the system. Lichess takes the opposite tactic of open security, where they not only openly publish their cheat detection methods, but encourage people to try and crack it so they can fix the flaws and make it stronger in response.
-
Danny Rensch explains how Chess.com's anti-cheat detection works (on Hikaru's stream)
Here's the lichess anticheat AI if anyone's interested. It's supposed to be a guide though, so when you report players it gives an estimate about whether they are probably a cheater or you were just mad about losing and reported and the cases it flags as cheaters are then dealt with by humans. Whatever algorithm you choose, you still need a human in there somewhere for moderation because false positives are a thing and for top players it could have terrible impact on their careers so you have to be sure.
-
The Chess World Isn’t Ready for a Cheating Scandal
Lichess is fully open-source and thus is their cheat-detection. They use https://github.com/clarkerubber/irwin which is a statistical model (tensorflow AI). This basically detects things such as using too much time on trivial moves such as forced captures, or other out-of-character moves.
maia-chess
-
Chess-GPT's Internal World Model
There is a very interesting project on this exact problem called Maia, which trains an engine based on millions of human games played on Lichess, specifically targeting varying levels of skill from 1300 to 1900 Elo. I haven't played it myself, by my understanding is that it does a much better job imitating the mistakes of human players. https://maiachess.com
- A chess terminal user interface implementation
-
Do you have to buy Maia to use it offline?
It's on GitHub, you just need an interface like Nibbler.
This Maia?
-
Anyone know if there are chess AIs trained like chatGPT, as a move predictor instead of a move maximizer as most have been (I think)?
That's exactly what Maia Chess is designed to do. https://maiachess.com/
Yes, https://maiachess.com/ is exactly what you're thinking of. It predicts the most likely next move at different ELO levels.
-
Making computer chess relevant to AI development... again?
I think that's what Maia is trying to do. It's a chess engine trained to predict human moves at specific skill levels.
-
Lichess: Over 2023, concurrent players online have been higher than the lockdown and Queen's Gambit boom; servers stable so far (@agadmator knows how to take us down!) - Check out all our features, 100% free to everyone: https://lichess.org/features no paywalls, no ads, 100% for love of chess
Lichess has Maia, which plays far more human-like than any bot on chess.com: https://maiachess.com/
-
The Q&A Megathread for new and beginner chess players
There are some community bots on lichess that use machine learning from a set of human games played on a certain level, like maia: https://maiachess.com/
-
Do you actually need to play humans to improve at chess?
Probably not nearly as much, but more than zero. See the Maia project for example.
What are some alternatives?
Stockfish - UCI chess engine
nnue-pytorch - Stockfish NNUE (Chess evaluation) trainer in Pytorch
stockfish - Integrates the Stockfish chess engine with Python
lichess-bot - A bridge between Lichess API and chess engines
Winter - UCI Chess Engine
lc0 - The rewritten engine, originally for tensorflow. Now all other backends have been ported here.
zahak - A UCI compatible chess AI in Go
Stockfish - A free and strong UCI chess engine
lila - ♞ lichess.org: the forever free, adless and open source chess server ♞
MongoDB - The MongoDB Database
python-chess-annotator - Reads chess games in PGN format and adds annotations using an engine
anarchychess-bot - The (un)official Lichess bot of r/AnarchyChess. Plays the Ruy Lopez, always captures en passant, never plays rook a4, and plays ke2!!/ke7!! when possible.