Leela Chess Zero Chess learned from scratch

Statement From LCZero Core Dev Team

Update from LCZero core devs team on DeusX TCEC entry

Two days before start of TCEC season 13, it was revealed that “DeusX engine” is just a set of NN weights trained by Albert Silver for the Lc0 engine.

The LCZero team had a very short period to analyze the situation and react to it, especially given that members of LCZero core dev team were in different timezones or on vacation, and given that even after the announcement it was not completely clear what the “DeusX engine” exactly was and which tools were used to produce the weights.

Because of this, and to avoid damaging the reputation of TCEC, we agreed to allow “DeusX engine” to participate in the TCEC.

Now that the dust has settled and we’ve had time to coordinate we would like to rectify a few things.

Albert Silver produced his set of the neural network weights using supervised learning. To do that, he needed incomparably less effort relative to the effort that he reused from the LCZero project, as the LCZero team had all the implementation work done, including building tools for the supervised training and the engine itself, defining neural network architecture etc. Given that, the work should NOT have been allowed into TCEC.

Albert Silver decided to hide information both from the LCZero community (by asking TCEC administration to be secret until the very last moment), and from TCEC administration (by not sharing appropriate information about importance of the engine vs neural network weights).

The team welcomes open and fair usage of the engine and tools around it, but in this case it was certainly not fair. We strongly condemn Albert’s submission which may be seen as plagiarism, and poor behavior.

  • The LCZero core developers.
Copyright (c) 2018 Dietrich Kappe