Efficiently updatable neural network

An efficiently updatable neural network (NNUE, a Japanese wordplay on Nue, sometimes stylised as ƎUИИ) is a neural network-based evaluation function that runs efficiently on central processing units without a requirement for a graphics processing unit (GPU). NNUE was invented by and introduced to computer shogi in 2018.^[1][2] On 6 August 2020, NNUE was for the first time ported to a chess engine, Stockfish 12.^[3][4] As of 2021, all of the top rated classical chess engines have an NNUE implementation to remain competitive.

NNUE is used primarily for the leaf nodes of the Alpha–beta tree.^[5] While being slower than traditional evaluation functions, NNUE does not suffer from the 'blindness beyond the current move' problem.^[6]

Compared to neural network evaluation based on dedicated GPUs, NNUE avoids idle times during the substantial data transfer operations between GPU and CPU required before and after each evaluation.^{[citation needed]}

The neural network used for shogi consists of four weight layers: W1 (16-bit integers) and W2, W3 and W4 (8-bit). Incremental computation and single instruction multiple data (SIMD) techniques are used with appropriate intrinsic instructions, specifically in the 2018 computer shogi implementation VPADDW, VPSUBW, VPMADDUBSW, VPACKSSDW, VPACKSSWB and VPMAXSB.^[1]

See also[]

• elmo (shogi engine)
• Stockfish chess engine

References[]

External links[]

• NNUE on the Chess Programming Wiki.
• NNUE evaluation functions for computer shogi on github.com