Efficiently updatable neural network

In computer strategy games like shogi and chess, an efficiently updatable neural network is a neural network-based evaluation function whose inputs are piece-square tables, or variants thereof like the king-piece-square table. NNUE relies on the tendency in these games for the game state to change only slightly between moves.
NNUE was invented by Yu Nasu and introduced to computer shogi in 2018. On 6 August 2020, NNUE was for the first time ported to a chess engine, Stockfish 12, resulting in a major increase in playing strength for that engine.
NNUE are designed to run efficiently on central processing units. They use incremental computation and single instruction multiple data techniques along with appropriate intrinsic instructions. In contrast, deep neural network-based chess engines such as Leela Chess Zero require GPU-based inference.
The neural network used for the original 2018 computer shogi implementation consists of four weight layers: W1 and W2, W3 and W4 . It has 4 fully-connected layers, ReLU activation functions, and outputs a single number, being the score of the board. As of 2025, Stockfish has introduced several optimizations to the NNUE architecture, but the overall architecture remains similar.