Sarit Khirirat
Sarit Khirirat
Division of Decision and Control, EECS, KTH Royal Institute of Technology
Verified email at kth.se - Homepage
Title
Cited by
Cited by
Year
The convergence of sparsified gradient methods
D Alistarh, T Hoefler, M Johansson, S Khirirat, N Konstantinov, C Renggli
arXiv preprint arXiv:1809.10505, 2018
1732018
Distributed learning with compressed gradients
S Khirirat, HR Feyzmahdavian, M Johansson
arXiv preprint arXiv:1806.06573, 2018
272018
Mini-batch gradient descent: Faster convergence under data sparsity
S Khirirat, HR Feyzmahdavian, M Johansson
2017 IEEE 56th Annual Conference on Decision and Control (CDC), 2880-2887, 2017
142017
Gradient compression for communication-limited convex optimization
S Khirirat, M Johansson, D Alistarh
2018 IEEE Conference on Decision and Control (CDC), 166-171, 2018
72018
Communication Efficient Sparsification for Large Scale Machine Learning
S Khirirat, S Magnússon, A Aytekin, M Johansson
arXiv preprint arXiv:2003.06377, 2020
22020
Convergence bounds for compressed gradient methods with memory based error compensation
S Khirirat, S Magnússon, M Johansson
ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and …, 2019
22019
Randomized first-order methods for convex optimization
S Khirirat
Master’s thesis, KTH Royal Institute of Technology, Stockholm, 2016
22016
Randomized first-order methods for convex optimization: Improved convergence rate bounds and experimental evaluations
S Khirirat
12016
Compressed Gradient Methods for Hessian-Aided Error Compensation
S Khirirat, S Magnússon, M Johansson
IEEE Transactions on Signal Processing, 2020
2020
A flexible framework for communication-efficient machine learning: from HPC to IoT
S Khirirat, S Magnússon, A Aytekin, M Johansson
arXiv e-prints, arXiv: 2003.06377, 2020
2020
First-Order Algorithms for Communication Efficient Distributed Learning
S Khirirat
KTH Royal Institute of Technology, 2019
2019
The system can't perform the operation now. Try again later.
Articles 1–11