Search found 4676 matches

by Ferdy
Fri Sep 17, 2021 1:05 pm
Forum: Computer Chess Club: General Topics
Topic: Poll test
Replies: 10
Views: 477

Re: Poll test

It works for me chrome, with ctrl+F5.
by Ferdy
Fri Sep 17, 2021 12:35 pm
Forum: Computer Chess Club: General Topics
Topic: Anti-cheating
Replies: 44
Views: 4420

Re: Anti-cheating

Took one Lichess user that violates TOS - terms of service. Get its rapid rating history and plot it with rolling window of 60 days as its rating is stable by more than 60 days. Confidence interval is +/- 2SD. Blue is actual rating, green is mean red is confidence interval. There was a sudden chang...
by Ferdy
Fri Sep 17, 2021 9:04 am
Forum: Computer Chess Club: General Topics
Topic: Anti-cheating
Replies: 44
Views: 4420

Re: Anti-cheating

Took one Lichess user that violates TOS - terms of service. Get its rapid rating history and plot it with rolling window of 60 days as its rating is stable by more than 60 days. Confidence interval is +/- 2SD. Blue is actual rating, green is mean red is confidence interval. There was a sudden change...
by Ferdy
Wed Sep 15, 2021 8:36 am
Forum: Computer Chess Club: General Topics
Topic: Anti-cheating
Replies: 44
Views: 4420

Re: Anti-cheating

Detecting anomalies in rating history. Plot the actual rating, take the mean and stdev from the rolling 14 days (take the mean in 14 days ...), plot the mean with confidence interval (CI) at twice the stdev. The blue is the actual rating, the green is the mean, the red covers the confidence interval...
by Ferdy
Tue Sep 14, 2021 7:04 pm
Forum: Computer Chess Club: General Topics
Topic: Anti-cheating
Replies: 44
Views: 4420

Re: Anti-cheating

With clock or time left info in the game notation, time per move or movetime can be calculated. 1. e4 { [%clk 0:03:00] } 1... e5 { [%clk 0:03:00] } 2. Nf3 { [%clk 0:02:56] } 2... Nc6 { [%clk 0:03:00] } ... Most novice cheater plays with almost constant movetime. By calculating the stdev, mean and CV...
by Ferdy
Tue Sep 14, 2021 5:43 am
Forum: Computer Chess Club: General Topics
Topic: Anti-cheating
Replies: 44
Views: 4420

Re: Anti-cheating

History is indeed useful, as generally the strength of players will stabilize at some point, and after that there could be improvements but only by a small increment, sudden change in the trend could be a suspect. Here is a plot where p2 player seems to be doing something. Not sure I understand all...
by Ferdy
Tue Sep 14, 2021 3:57 am
Forum: Computer Chess Club: General Topics
Topic: Yet another hparameter tuner called Mainit
Replies: 1
Views: 457

Re: Yet another hparameter tuner called Mainit

After 100 trials are completed, top trial is 44 with 79 elo value.

Image
by Ferdy
Tue Sep 14, 2021 3:40 am
Forum: Computer Chess Club: General Topics
Topic: Anti-cheating
Replies: 44
Views: 4420

Re: Anti-cheating

I think the interesting idea is to train a neural network to detect engine/human games. After you train it on a large enough dataset, it will be relatively easy to check statistics for false positives for such a network. Well, that's exactly what they did for lichess. https://github.com/clarkerubbe...
by Ferdy
Mon Sep 13, 2021 6:35 pm
Forum: Computer Chess Club: General Topics
Topic: Anti-cheating
Replies: 44
Views: 4420

Re: Anti-cheating

Make a model with keras and TF from random data, with basic features such as mean error, blunder counts and others and make the model predict on some data after training. Generate 1000 random data points, there are 50 samples below. meanerror blunder mistake dubious top1pct cheat 0 166 1 14 2 25.0 0...
by Ferdy
Mon Sep 13, 2021 4:24 pm
Forum: Computer Chess Club: General Topics
Topic: Yet another hparameter tuner called Mainit
Replies: 1
Views: 457

Yet another hparameter tuner called Mainit

This system is using microsoft nni - neural network intelligence toolkit. It has some tuners to use. Sample overview dashboard. The default metric column is the Elo of the parameter values under the trial. https://i.imgur.com/ws6z6KZ.png Trial details metric, figure is Elo as objective value after e...