Pith. sign in

ACM Transactions on Architecture and Code Optimization (TACO) (2024)

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

citation-role summary

background 1

citation-polarity summary

fields

cs.LG 1

years

2024 1

verdicts

CONDITIONAL 1

roles

background 1

polarities

unclear 1

representative citing papers

DQA: An Efficient Method for Deep Quantization of Deep Neural Network Activations

cs.LG · 2024-12-12 · conditional · novelty 5.0

DQA quantizes neural network activations to below 6 bits by using extra precision on important channels, truncating it with bit shifts, and Huffman-coding the truncation error, improving accuracy over direct quantization and NoisyQuant in most, but not all, tested settings.

citing papers explorer

Showing 1 of 1 citing paper.

  • DQA: An Efficient Method for Deep Quantization of Deep Neural Network Activations cs.LG · 2024-12-12 · conditional · none · ref 6

    DQA quantizes neural network activations to below 6 bits by using extra precision on important channels, truncating it with bit shifts, and Huffman-coding the truncation error, improving accuracy over direct quantization and NoisyQuant in most, but not all, tested settings.