RATQ: A universal fixed-length quantizer for stochastic optimization

P Mayekar, H Tyagi - International Conference on Artificial …, 2020 - proceedings.mlr.press
Abstract We present Rotated Adaptive Tetra-iterated Quantizer (RATQ), afixed-length
quantizer for gradients in first order stochasticoptimization. RATQ is easy to implement and …
被引用次数：45 相关文章

RATQ: A Universal Fixed-Length Quantizer for Stochastic Optimization

P Mayekar, H Tyagi - arXiv e-prints, 2019 - ui.adsabs.harvard.edu
Abstract We present Rotated Adaptive Tetra-iterated Quantizer (RATQ), a fixed-length
quantizer for gradients in first order stochastic optimization. RATQ is easy to implement and …

RATQ: A Universal Fixed-Length Quantizer for Stochastic Optimization

P Mayekar, H Tyagi - arXiv preprint arXiv:1908.08200, 2019 - arxiv.org
We present Rotated Adaptive Tetra-iterated Quantizer (RATQ), a fixed-length quantizer for
gradients in first order stochastic optimization. RATQ is easy to implement and involves only …
[PDF] github.io

[PDF][PDF] RATQ: A Universal Fixed-Length Quantizer for Stochastic Optimization

P Mayekar, H Tyagi - prathamesh220.github.io
Abstract We present Rotated Adaptive Tetra-iterated Quantizer (RATQ), a fixed-length
quantizer for gradients in first order stochastic optimization. RATQ is easy to implement and …
[PDF] iisc.ac.in

[PDF][PDF] RATQ: A Universal Fixed-Length Quantizer for Stochastic Optimization

P Mayekar, H Tyagi - ece.iisc.ac.in
Abstract We present Rotated Adaptive Tetra-iterated Quantizer (RATQ), a fixed-length
quantizer for gradients in first order stochastic optimization. RATQ is easy to implement and …

RATQ: A Universal Fixed-Length Quantizer for Stochastic Optimization

P Mayekar, H Tyagi - International Conference on Artificial …, 2020 - proceedings.mlr.press
Abstract We present Rotated Adaptive Tetra-iterated Quantizer (RATQ), afixed-length
quantizer for gradients in first order stochasticoptimization. RATQ is easy to implement and …

RATQ: A Universal Fixed-Length Quantizer for Stochastic Optimization

P Mayekar, H Tyagi - IEEE Transactions on Information Theory, 2021 - ieeexplore.ieee.org
We present Rotated Adaptive Tetra-iterated Quantizer (RATQ), a fixed-length quantizer for
gradients in first order stochastic optimization. RATQ is easy to implement and involves only …
[PDF] github.io

[PDF][PDF] RATQ: A Universal Fixed-Length Quantizer for Stochastic Optimization

P Mayekar, H Tyagi - prathamesh220.github.io
RATQ: A Universal Fixed-Length Quantizer for Stochastic Optimization Page 1 RATQ: A
Universal Fixed-Length Quantizer for Stochastic Optimization Prathamesh Mayekar Joint work …