Follow
Brian Chmiel
Brian Chmiel
PhD, AI Research Scientist @ Intel-HabanaLabs
Verified email at habana.ai
Title
Cited by
Cited by
Year
Loss aware post-training quantization
Y Nahshan, B Chmiel, C Baskin, E Zheltonozhskii, R Banner, ...
Machine Learning 110 (11), 3245-3262, 2021
1732021
Accelerated sparse neural training: A provable and efficient method to find n: m transposable masks
I Hubara, B Chmiel, M Island, R Banner, J Naor, D Soudry
Advances in neural information processing systems 34, 21099-21111, 2021
1012021
Robust quantization: One model to rule them all
B Chmiel, R Banner, G Shomron, Y Nahshan, A Bronstein, U Weiser
Advances in neural information processing systems 33, 5308-5317, 2020
812020
Neural gradients are near-lognormal: improved quantized and sparse training
B Chmiel, L Ben-Uri, M Shkolnik, E Hoffer, R Banner, D Soudry
arXiv preprint arXiv:2006.08173, 2020
502020
Feature map transform coding for energy-efficient CNN inference
B Chmiel, C Baskin, E Zheltonozhskii, R Banner, Y Yermolin, ...
2020 International Joint Conference on Neural Networks (IJCNN), 1-9, 2020
272020
Logarithmic unbiased quantization: Simple 4-bit training in deep learning
B Chmiel, R Banner, E Hoffer, HB Yaacov, D Soudry
arXiv preprint arXiv:2112.10769, 2021
152021
Optimal fine-grained n: M sparsity for activations and neural gradients
B Chmiel, I Hubara, R Banner, D Soudry
arXiv preprint arXiv:2203.10991, 2022
122022
CAT: Compression-Aware Training for bandwidth reduction
C Baskin, B Chmiel, E Zheltonozhskii, R Banner, AM Bronstein, ...
Journal of Machine Learning Research 22 (269), 1-20, 2021
122021
Logarithmic unbiased quantization: Practical 4-bit training in deep learning
B Chmiel, R Banner, E Hoffer, HB Yaacov, D Soudry
112021
Accurate neural training with 4-bit matrix multiplications at standard formats
B Chmiel, R Banner, E Hoffer, H Ben-Yaacov, D Soudry
The Eleventh International Conference on Learning Representations, 2023
82023
Colored noise injection for training adversarially robust neural networks
E Zheltonozhskii, C Baskin, Y Nemcovsky, B Chmiel, A Mendelson, ...
arXiv preprint arXiv:2003.02188, 2020
72020
Neural gradients are lognormally distributed: understanding sparse and quantized training
B Chmiel, L Ben-Uri, M Shkolnik, E Hoffer, R Banner, D Soudry
arXiv, 2020
62020
Towards learning of filter-level heterogeneous compression of convolutional neural networks
Y Zur, C Baskin, E Zheltonozhskii, B Chmiel, I Evron, AM Bronstein, ...
arXiv preprint arXiv:1904.09872, 2019
62019
Adversarial robustness via noise injection in smoothed models
Y Nemcovsky, E Zheltonozhskii, C Baskin, B Chmiel, AM Bronstein, ...
Applied Intelligence 53 (8), 9483-9498, 2023
52023
Minimum variance unbiased n: M sparsity for the neural gradients
B Chmiel, I Hubara, R Banner, D Soudry
The Eleventh International Conference on Learning Representations, 2023
52023
Bimodal-distributed binarized neural networks
T Rozen, M Kimhi, B Chmiel, A Mendelson, C Baskin
Mathematics 10 (21), 4107, 2022
52022
Smoothed inference for adversarially-trained models
Y Nemcovsky, E Zheltonozhskii, C Baskin, B Chmiel, M Fishman, ...
arXiv preprint arXiv:1911.07198, 2019
22019
EXAQ: Exponent Aware Quantization For LLMs Acceleration
M Shkolnik, M Fishman, B Chmiel, H Ben-Yaacov, R Banner, KY Levy
arXiv preprint arXiv:2410.03185, 2024
2024
Scaling FP8 training to trillion-token LLMs
M Fishman, B Chmiel, R Banner, D Soudry
arXiv preprint arXiv:2409.12517, 2024
2024
Compressing neural networks through unbiased minimum variance pruning
B Chmiel, I Hubara, R Banner
US Patent App. 18/164,875, 2024
2024
The system can't perform the operation now. Try again later.
Articles 1–20