Winter, an UCI compliant open source chess engine by Jonathan Rosenthal, written in C++, released on January 08, 2018 under the terms of GPL Version 3. Winter is inspired by machine learning techniques, as applied in move ordering and in particular in evaluation, and heavily relies on C++ templates but not on any library aside from STL as it comes with its own implementations for statistics [1] and linear algebra [2]. Winter started its life in 2016 as a group project at ETH Zurich in a course on parallel computing along with Jonas Kuratli and Jonathan Maurer - the current release with Jonathan Rosenthal as sole author has removed the parallel portion of the code [3]. It started to play on-line at HGM's Online Engine Blitz Tourneys in April 2017 [4].
Winter in Zurich, ETH Zurich view [5]

Selected Features

Board Representation

• Bitboards
• 8x8 Board
• Fancy Magic Bitboards
  BMI2 - PEXT Bitboards

Search

• Fail-Hard Iterative Deepening
• Alpha-Beta Principal Variation Search
• Transposition Table
• Move Ordering is based on the linear part of a Logistic Regression classifier aka cluster analysis
  • The classifier is trained via temporal difference learning to predict whether a move will return beta
  • Classifier considers TT move, killers, move type, from square, target square, capture target, SEE, target square of last move, check and changes between forcing and unforcing moves (a capture is more likely after another capture)
• Selectivity
  • Null Move Reductions
  • Static Null Move Pruning
  • Late Move Reductions
  • Futility Pruning
  • Quiescence Search
  • Static Exchange Evaluation

Evaluation

• Non standard approach relies on a mixture model ^[7]
  • Assumes positions encountered in search come from some set of k Gaussians ^{[8] [9]}

    A set of Gaussians

  • Mixture model is trained via EM algorithm ^{[10] [11]} either on database games or positions sampled from search
  • For each Gaussian a separate evaluation function is trained. When the evaluation function is called the relative probability a position stems from each Gaussian is estimated, the evaluation functions are computed and the final score is returned as the weighted average - a generalization of tapered eval with game phases ^[12]
• Parameter weights are trained via a mixture of reinforcement (temporal difference) learning and supervised learning
  • Minimizing the cross entropy loss of a Logistic Regression model for each of the k Gaussians
  • Training converges fast due to linear model at the heart

Misc

• Perft
• Print pretty Unicode chess boards

Forum Posts

• Re: Tapered Eval between 4 phases by Jonathan Rosenthal, CCC, October 16, 2017 » Tapered Eval
• Winter Released by Jonathan Rosenthal, CCC, January 08, 2018
  Re: Winter Released by Jonathan Rosenthal, CCC, January 09, 2018
  Windows version released by Jonathan Rosenthal, CCC, January 23, 2018

External Links

Chess Engine

• GitHub - rosenthj/Winter: UCI Chess Engine

Misc

• Winter from Wikipedia
• Winter - Wiktionary
• winter - Wiktionary
• Winter (disambiguation) from Wikipedia
• Summer and Winter Schools | ETH Zurich
• Wintel from Wikipedia
• AI winter from Wikipedia
• Winter (surname) from Wikipedia
• Edward Winter (chess historian) from Wikipedia
• Johnny Winter - Winter Ballade, ina.fr 1970, YouTube Video
  feat. Tommy Shannon on bass and Uncle John Turner on drums
  

References

1. ^ Winter/statistics.h at master · rosenthj/Winter · GitHub
2. ^ Winter/linear_algebra.h at master · rosenthj/Winter · GitHub
3. ^ Winter Released by Jonathan Rosenthal, CCC, January 08, 2018
4. ^ Re: On-line engine blitz tourney April by Harm Geert Muller, CCC, April 22, 2017
5. ^ Image clipped from Winter views from our floor by Bertrand Meyer, ETH Zurich, Gray moment with Zentralbibliothek and Predigerkirche in the foreground, Lake Zurich, Grossmünster, Fraumünster and St. Peter behind, in the background foothill of Uetliberg, from Bertrand Meyer's ETH home page
6. ^ based on Winter Released by Jonathan Rosenthal, CCC, January 08, 2018
7. ^ Re: Winter Released by Jonathan Rosenthal, CCC, January 09, 2018
8. ^ k-means clustering from Wikipedia
9. ^ K Means by Chris Piech
10. ^ The EM Algorithm for Gaussian Mixtures - Probabilistic Learning: Theory and Algorithms, CS 274A (pdf) University of California, Irvine
11. ^ Mixture Models & EM algorithm Lecture 21 (pdf) by David Sontag, New York University
12. ^ Re: Tapered Eval between 4 phases by Jonathan Rosenthal, CCC, October 16, 2017

What links here?