 Meta-Learning with a Geometry-Adaptive PreconditionerComputer Vision and Pattern Recognition (CVPR), 2023 |
 SparseProp: Efficient Sparse Backpropagation for Faster Training of
Neural NetworksInternational Conference on Machine Learning (ICML), 2023 |
 AskewSGD : An Annealed interval-constrained Optimisation method to train
Quantized Neural NetworksInternational Conference on Artificial Intelligence and Statistics (AISTATS), 2022 |
 Accurate Neural Training with 4-bit Matrix Multiplications at Standard
FormatsInternational Conference on Learning Representations (ICLR), 2021 |
 L2ight: Enabling On-Chip Learning for Optical Neural Networks via
Efficient in-situ Subspace OptimizationNeural Information Processing Systems (NeurIPS), 2021 |
 No frame left behind: Full Video Action RecognitionComputer Vision and Pattern Recognition (CVPR), 2021 |
 DeepCABAC: A Universal Compression Algorithm for Deep Neural NetworksIEEE Journal on Selected Topics in Signal Processing (JSTSP), 2019 |
