Parallel Restarted SGD with Faster Convergence and Less Communication: Demystifying Why Model Averaging Works for Deep Learning H Yu, S Yang, S Zhu
Proceedings of the AAAI Conference on Artificial Intelligence 33, 5693-5700, 2019
627 * 2019 On the linear speedup analysis of communication efficient momentum sgd for distributed non-convex optimization H Yu, R Jin, S Yang
International Conference on Machine Learning, 7184-7193, 2019
366 2019 Online convex optimization with stochastic constraints H Yu, M Neely, X Wei
Advances in Neural Information Processing Systems 30, 2017
214 2017 Online convex optimization with time-varying constraints MJ Neely, H Yu
arXiv preprint arXiv:1702.04783, 2017
104 2017 A low complexity algorithm with O (√ T) regret and O (1) constraint violations for online convex optimization with long term constraints H Yu, MJ Neely
Journal of Machine Learning Research 21 (1), 1-24, 2020
83 * 2020 A Simple Parallel Algorithm with an Convergence Rate for General Convex Programs H Yu, MJ Neely
SIAM Journal on Optimization 27 (2), 759-783, 2017
59 2017 Online primal-dual mirror descent under stochastic constraints X Wei, H Yu, MJ Neely
Proceedings of the ACM on Measurement and Analysis of Computing Systems 4 (2 …, 2020
57 2020 On the Computation and Communication Complexity of Parallel SGD with Dynamic Batch Sizes for Stochastic Non-Convex Optimization H Yu, R Jin
International Conference on Machine Learning, 7174-7183, 2019
55 2019 Rank-constrained Schur-convex optimization with multiple trace/log-det constraints H Yu, VKN Lau
IEEE transactions on signal processing 59 (1), 304-314, 2010
38 2010 Learning-aided optimization for energy-harvesting devices with outdated state information H Yu, MJ Neely
IEEE/ACM Transactions on Networking 27 (4), 1501-1514, 2019
33 2019 A new backpressure algorithm for joint rate control and routing with vanishing utility optimality gaps and finite queue lengths H Yu, MJ Neely
IEEE/ACM Transactions on Networking 26 (4), 1605-1618, 2018
29 2018 Online learning in weakly coupled markov decision processes: A convergence time study X Wei, H Yu, MJ Neely
Proceedings of the ACM on Measurement and Analysis of Computing Systems 2 (1 …, 2018
19 2018 A primal-dual type algorithm with the O (1/t) convergence rate for large scale constrained convex programs H Yu, MJ Neely
2016 IEEE 55th Conference on Decision and Control (CDC), 1900-1905, 2016
18 2016 Solving Non-smooth Constrained Programs with Lower Complexity than : A Primal-Dual Homotopy Smoothing Approach X Wei, H Yu, Q Ling, M Neely
Advances in neural information processing systems 31, 2018
17 2018 A low complexity algorithm with O (√ T) regret and finite constraint violations for online convex optimization with long term constraints H Yu, MJ Neely
arXiv preprint arXiv:1604.02218, 2016
17 2016 Duality codes and the integrality gap bound for index coding H Yu, MJ Neely
IEEE Transactions on Information Theory 60 (11), 7256-7268, 2014
16 2014 Dynamic transmit covariance design in MIMO fading systems with unknown channel distributions and inaccurate channel state information H Yu, MJ Neely
IEEE Transactions on Wireless Communications 16 (6), 3996-4008, 2017
15 2017 On the convergence time of the drift-plus-penalty algorithm for strongly convex programs H Yu, MJ Neely
2015 54th IEEE Conference on Decision and Control (CDC), 2673-2679, 2015
15 2015 On the convergence time of dual subgradient methods for strongly convex programs H Yu, MJ Neely
IEEE Transactions on Automatic Control 63 (4), 1105-1112, 2017
14 2017 A primal-dual parallel method with O (1/ϵ) convergence for constrained composite convex programs H Yu, MJ Neely
arXiv preprint arXiv:1708.00322 7, 2017
11 2017