Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
QScheduler: Adaptive Gradient Sampling for Zeroth-Order On-Device Training on INT8 NPUs
Victor Felipe Domingues Do Amaral, Pierre Demaj, Erwan Libessart +3
Zeroth-Order (ZO) optimization enables On-Device Learning (ODL) on NPU-equipped microcontrollers by estimating gradients through forward passes alone, bypassing the need for backpr…
cs.LG2023
Neural network scoring for efficient computing
Hugo Waltsburger, Erwan Libessart, Chengfang Ren +2
Much work has been dedicated to estimating and optimizing workloads in high-performance computing (HPC) and deep learning. However, researchers have typically relied on few metrics…