Ron Banner
14 papers · 2018–2025 · 5 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+6 more ↓ Show less ↑
π Interdisciplinary Bridge πΊοΈ Taxonomy Completionist (13) π Academic Marathon (7) π Conference Polyglot (5) π§ Keyword Pioneer
π
Conference Polyglot
(5)
π
Academic Marathon
(7)
π
Triple Crown
π€
Dynamic Duo
(10)
β‘
Prolific Year
(5)
π
Century Club
(14)
Conferences
NIPS (6)
ICLR (5)
ECCV (1)
ICML (1)
JMLR (1)
Top co-authors
Keywords
model compression
(4)
neural network quantization
(4)
batch normalization
(2)
post-training quantization
(2)
convolutional neural network
(1)
convolutional network
(1)
weight decay
(1)
numerical stability
(1)
structural pruning
(1)
quantization error
(1)
weight normalization
(1)
matrix multiplication
(1)
memory bandwidth
(1)
deep network
(1)
gradient quantization
(1)
bit-width allocation
(1)
4-bit quantization
(1)
bit allocation
(1)
bandwidth reduction
(1)
calibration set
(1)
Papers
Scaling FP8 training to trillion-token LLMs
ICLR 2025
DropCompute: simple and more robust distributed synchronous training via compute variance reduction
NIPS 2023
Minimum Variance Unbiased N:M Sparsity for the Neural Gradients
ICLR 2023
Accurate Neural Training with 4-bit Matrix Multiplications at Standard Formats
ICLR 2023
CAT: Compression-Aware Training for bandwidth reduction
JMLR 2021
Accelerated Sparse Neural Training: A Provable and Efficient Method to Find N:M Transposable Masks
NIPS 2021
Neural gradients are near-lognormal: improved quantized and sparse training
ICLR 2021
GAN "Steerability" without optimization
ICLR 2021
Accurate Post Training Quantization With Small Calibration Sets
ICML 2021
Robust Quantization: One Model to Rule Them All
NIPS 2020
Thanks for Nothing: Predicting Zero-Valued Activations with Lightweight Convolutional Neural Networks
ECCV 2020
Post training 4-bit quantization of convolutional networks for rapid-deployment
NIPS 2019
Scalable methods for 8-bit training of neural networks
NIPS 2018
Norm matters: efficient and accurate normalization schemes in deep networks
NIPS 2018