Publications

Sarit Khirirat, Xiaoyu Wang, Sindri Magnússon, Mikael Johansson (2023). Improved Step-Size Schedules for Proximal Noisy Gradient Methods. IEEE Transactions on Signal Processing (Accepted) 2023.

Sarit Khirirat (2022). First-Order Algorithms for Communication Efficient Distributed Learning. PhD Thesis:: Stockholm: KTH Royal Institute of Technology, 2022. , p. 233.

Erik Berglund, Sarit Khirirat, Xiaoyu Wang (2022). Zeroth-order randomized subspace Newton methods. ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

Sarit Khirirat, Sindri Magnússon, Mikael Johansson (2022). Eco-Fedsplit: Federated Learning with Error-Compensated Compression. ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

Sarit Khirirat, Sindri Magnússon, Arda Aytekin, Mikael Johansson (2021). A Flexible Framework for Communication-Efficient Machine Learning. Proceedings of the AAAI Conference on Artificial Intelligence.

Sarit Khirirat, Sindri Magnússon, Mikael Johansson (2020). Compressed Gradient Methods With Hessian-Aided Error Compensation. IEEE Transactions on Signal Processing ( Volume: 69), 2020, p. 998 - 1011.

Sarit Khirirat (2019). First-Order Algorithms for Communication Efficient Distributed Learning. Licentiate Thesis:: Stockholm: KTH Royal Institute of Technology, 2019. , p. 106.

Sarit Khirirat, Sindri Magnússon, Mikael Johansson (2019). Convergence Bounds for Compressed Gradient Methods with Memory Based Error Compensation. ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

Dan Alistarh, Torsten Hoefler, Mikael Johansson, Nikola Konstantinov, Sarit Khirirat, Cédric Renggli (2018). The convergence of sparsified gradient methods. Advances in Neural Information Processing Systems.

Sarit Khirirat, Mikael Johansson, Dan Alistarh (2018). Gradient compression for communication-limited convex optimization. 2018 IEEE Conference on Decision and Control (CDC).

Sarit Khirirat, Hamid Reza Feyzmahdavian, Mikael Johansson (2018). Distributed learning with compressed gradients. arXiv preprint arXiv:1806.06573.

Sarit Khirirat, Hamid Reza Feyzmahdavian, Mikael Johansson (2017). Mini-batch gradient descent: Faster convergence under data sparsity. 2017 IEEE 56th Annual Conference on Decision and Control (CDC).