3 papers
cs.DC2026
Speeding up Log-Sum-Exp: Kernel Fusion at the Memory Wall, Integer Arithmetic at the Compute Wall
Lingyun Yao, Martin Andraud, Niki Loppi +4
While the Log-Sum-Exp (LSE) function is a foundational numerical primitive for stable log-domain computation in modern machine learning, its fast and efficient computation on gener…
cs.LG2025
Hardware-efficient tractable probabilistic inference for TinyML Neurosymbolic AI applications
Jelin Leslin, Martin Trapp, Martin Andraud
Neurosymbolic AI (NSAI) has recently emerged to mitigate limitations associated with deep learning (DL) models, e.g. quantifying their uncertainty or reason with explicit rules. He…
cs.LG2024
On Hardware-efficient Inference in Probabilistic Circuits
Lingyun Yao, Martin Trapp, Jelin Leslin +4
Probabilistic circuits (PCs) offer a promising avenue to perform embedded reasoning under uncertainty. They support efficient and exact computation of various probabilistic inferen…