Method and system for training binary quantized weight and activation function for deep neural networks LI Xinlin, S Darabi, M Belbahri, VP NIA
US Patent App. 16/582,131, 2020
21 2020 BinaryViT: pushing binary vision transformers towards convolutional models PHC Le, X Li
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern …, 2023
6 2023 S : Sign-Sparse-Shift Reparametrization for Effective Training of Low-bit Shift Networks X Li, B Liu, Y Yu, W Liu, C Xu, V Partovi Nia
Advances in Neural Information Processing Systems 34, 14555-14566, 2021
6 2021 Euclidnets: An alternative operation for efficient inference of deep learning models X Li, M Parazeres, A Oberman, A Ghaffari, M Asgharian, VP Nia
SN Computer Science 4 (5), 507, 2023
4 2023 Deep neural networks pruning via the structured perspective regularization M Cacciola, A Frangioni, X Li, A Lodi
SIAM Journal on Mathematics of Data Science 5 (4), 1051-1077, 2023
2 2023 Low-bit Shift Network for End-to-End Spoken Language Understanding AR Avila, K Bibi, RH Yang, X Li, C Xing, X Chen
Interspeech 2022, 2022
2 2022 DenseShift: Towards Accurate and Efficient Low-Bit Power-of-Two Quantization X Li, B Liu, RH Yang, V Courville, C Xing, VP Nia
Proceedings of the IEEE/CVF International Conference on Computer Vision …, 2023
1 * 2023 Methods, systems, and media for low-bit neural networks using bit shift operations LI Xinlin, VP NIA
US Patent App. 18/521,425, 2024
2024 Understanding Neural Network Binarization with Forward and Backward Proximal Quantizers Y Lu, Y Yu, X Li, V Partovi Nia
Advances in Neural Information Processing Systems 36, 2024
2024 Methods and systems for computing an output of a neural network layer LI Xinlin, MO Prazeres, AM Oberman, VP NIA
US Patent App. 17/317,833, 2022
2022 A causal direction test for heterogeneous populations VP Nia, X Li, M Asgharian, S Hu, Y Geng, Z Chen
Machine Learning with Applications 7, 100235, 2022
2022 Low resource computational block for a trained neural network LI Xinlin, VP NIA
US Patent App. 16/900,658, 2021
2021