Pith. sign in

REVIEW 1 cited by

ReCU: Reviving the Dead Weights in Binary Neural Networks

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2103.12369 v2 pith:K3H4Q42X submitted 2021-03-23 cs.LG cs.CV

classification cs.LGcs.CV
keywords weightsdeadbnnsrecuerrorquantizationthentraining
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Binary neural networks (BNNs) have received increasing attention due to their superior reductions of computation and memory. Most existing works focus on either lessening the quantization error by minimizing the gap between the full-precision weights and their binarization or designing a gradient approximation to mitigate the gradient mismatch, while leaving the "dead weights" untouched. This leads to slow convergence when training BNNs. In this paper, for the first time, we explore the influence of "dead weights" which refer to a group of weights that are barely updated during the training of BNNs, and then introduce rectified clamp unit (ReCU) to revive the "dead weights" for updating. We prove that reviving the "dead weights" by ReCU can result in a smaller quantization error. Besides, we also take into account the information entropy of the weights, and then mathematically analyze why the weight standardization can benefit BNNs. We demonstrate the inherent contradiction between minimizing the quantization error and maximizing the information entropy, and then propose an adaptive exponential scheduler to identify the range of the "dead weights". By considering the "dead weights", our method offers not only faster BNN training, but also state-of-the-art performance on CIFAR-10 and ImageNet, compared with recent methods. Code can be available at https://github.com/z-hXu/ReCU.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Dedicated Inference Engine and Binary-Weight Neural Networks for Lightweight Instance Segmentation

    cs.CV 2025-01 conditional novelty 4.0 of 10

    A dual-mode XNOR/AND bitwise MAC architecture for 8-bit activation, 1-bit weight inference reduces gate count to about 52% of prior designs, alongside binary-weight instance segmentation networks that claim higher Per...

Pith tools