<ul><li>BMRS (Bayesian Model Reduction for Structured Pruning) is a fully end-to-end Bayesian method of structured pruning.</li><li>It is based on Bayesian structured pruning with multiplicative noise and Bayesian model reduction (BMR), allowing efficient comparison of Bayesian models under a change in prior.</li><li>The two realizations of BMRS, BMRS_N and BMRS_U, derived from different priors, offer reliable compression rates and accuracy without the need for tuning thresholds, and achieve aggressive compression based on truncation boundaries, respectively.</li><li>Experiments on multiple datasets and neural networks showed that BMRS provides a competitive performance-efficiency trade-off compared to other pruning methods.</li></ul>

BMRS: Bayesian Model Reduction for Structured Pruning

Discover more