2023/11/16 by Jannis Schönleber, Schönleber, Jannis, Lukas Cavigelli +7 · 1 voice · 2 citations
Computer Science · Mathematics · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Hardware Architecture (cs.AR) #Machine Learning (cs.LG) #Machine Learning (stat.ML) #cs.AR #cs.CV #cs.LG #stat.ML
paper · pdf · doi:10.48550/arxiv.2311.10207
arxiv published 2023/11/16 · arxiv updated 2025/07/25
Artificial intelligence has surged in recent years, with advancements in machine learning rapidly impacting nearly every area of life. However, the growing complexity of these models has far outpaced advancements in available hardware accelerators, leading to significant computational and energy demands, primarily due to matrix multiplications, which dominate the compute workload. Maddness (i.e., Multiply-ADDitioN-lESS) presents a hash-based version of product quantization, which renders matrix multiplications into lookups and additions, eliminating the need for multipliers entirely. We present Stella Nera, the first Maddness-based accelerator achieving an energy efficiency of 161 TOp/s/[email protected], 25x better than conventional MatMul accelerators due to its small components and reduced computational complexity. We further enhance Maddness with a differentiable approximation, allowing for gradient-based fine-tuning and achieving an end-to-end performance of 92.5% Top-1 accuracy on CIFAR-10.