← Search

Kazushi Kawamura

2 accepted papers

2025

Binary Quadratic Quantization: Beyond First-Order Quantization for Real-Valued Matrix Compression

NeurIPS 2025poster

This paper proposes a novel matrix quantization method, Binary Quadratic Quan- tization (BQQ). In contrast to conventional first-order quantization approaches— such as uniform quantization and binary coding quantization—that approximate real-valued matrices via linear combinations of binary bases, B…

Cited by 0Source ↗Scholar ↗
2022

Multicoated Supermasks Enhance Hidden Networks

ICML 2022spotlight

Hidden Networks (Ramanujan et al., 2020) showed the possibility of finding accurate subnetworks within a randomly weighted neural network by training a connectivity mask, referred to as supermask. We show that the supermask stops improving even though gradients are not zero, thus underutilizing back…

Cited by 13Source ↗PDF ↗Scholar ↗