- arXivUnmasking the Lottery Ticket Hypothesis: What’s Encoded in a Winning Ticket’s Mask?arXiv preprint arXiv:2210.03044, 2022
- NeurIPSLottery Tickets on a Data Diet: Finding Initializations with Sparse Trainable Networks36th Conference on Neural Information Processing Systems, 2022Spotlight presentation at the Sparsity in Neural Networks (SNN) Workshop 2022.
- SIMAXPractical leverage-based sampling for low-rank tensor decompositionSIAM Journal on Matrix Analysis and Applications (SIMAX), 2022
- PLOS Comp BioTowards a more general understanding of the algorithmic utility of recurrent connectionsPLOS Computational Biology, 2022Poster presentation at COSYNE 2019.
- ICLRHow many degrees of freedom do we need to train deep networks: a loss landscape perspectiveInternational Conference on Learning Representations, 2022
- arXivAvoiding Spurious Local Minima in Deep Quadratic NetworksarXiv preprint arXiv:2001.00098, 2020Oral presentation at Deep Math 2019.
- ThesisGraph-Based Clustering: Distributed Algorithms for Balanced Graph Cuts2016
- SPIEApplying matching pursuit decomposition time-frequency processing to UGS footstep classificationIn Sensors, and Command, Control, Communications, and Intelligence (C3I) Technologies for Homeland Security and Homeland Defense XII 2013