DigiNews

Tech Watch Articles

← Back to articles

Better activation functions for NNUE

Quality: 8/10 Relevance: 9/10

Summary

An experimental study of activation functions for Viridithas's NNUE shows replacing SCReLU layers with Swish, and then SwiGLU on L2, yields Elo gains and smoother evaluation scales. The article discusses challenges with Hard-Swish causing reduced sparsity, the use of regularization to restore density, and the broader potential of applying DL insights to chess NNUE design.

🚀 Service construit par Johan Denoyer