Showing cs.LGShow all
2 papers · 1 filter
cs.LG2025
Frugality in second-order optimization: floating-point approximations for Newton's method
Giuseppe Carrino, Elena Loli Piccolomini, Elisa Riccietti +1
Minimizing loss functions is central to machine-learning training. Although first-order methods dominate practical applications, higher-order techniques such as Newton's method can…
cs.LG2025
Mixed precision accumulation for neural network inference guided by componentwise forward error analysis
El-Mehdi El Arar, Silviu-Ioan Filip, Theo Mary +1
This work proposes a mathematically founded mixed precision accumulation strategy for the inference of neural networks. Our strategy is based on a new componentwise forward error a…