Publications

2024

Small steps no more: Global convergence of stochastic gradient bandits for arbitrary learning rates
Jincheng Mei, Bo Dai, Alekh Agarwal, Sharan Vaswani, Anant Raj, Csaba Szepesvari, Dale Schuurmans, NeurIPS, 2024. [pdf]

Fast Convergence of Softmax Policy Mirror Ascent
Reza Asad, Reza Babanezhad, Issam Laradji, Nicolas Le Roux, Sharan Vaswani, "Optimization for Machine Learning" workshop, NeurIPS, 2024. [pdf]

Improving OOD Generalization of Pre-trained Encoders via Aligned Embedding-Space Ensembles
Shuman Peng, Arash Khoeini, Sharan Vaswani, Martin Ester, Workshop on Unifying Representations in Neural Models, NeurIPS, 2024. [pdf]

From Inverse Optimization to Feasibility to ERM
Saurabh Mishra, Anant Raj, Sharan Vaswani , ICML, 2024. [pdf] [code] [video]
"Optimization for Machine Learning" workshop, NeurIPS, 2023.

Towards Principled, Practical Policy Gradient for Bandits and Tabular MDPs
Michael Lu, Matin Aghaei, Anant Raj, Sharan Vaswani, RLC, 2024. [pdf] [code] [slides][Talk at Alberta]
"Optimization for Machine Learning" workshop, NeurIPS, 2023 (Oral Presentation)

(Accelerated) Noise-adaptive Stochastic Heavy-Ball Momentum
Anh Dang, Reza Babanezhad, Sharan Vaswani [pdf] [code]
"Optimization for Machine Learning" workshop, NeurIPS, 2023.

2023

Decision-Aware Actor-Critic with Function Approximation and Theoretical Guarantees
Sharan Vaswani, Amirreza Kazemi, Reza Babanezhad, Nicolas Le Roux, NeurIPS, 2023. [pdf] [slides] [Talk at Vector]
“Duality Principles for Modern ML” workshop, ICML, 2023.

Surrogate Minimization: An Optimization Algorithm for Training Large Neural Networks with Model Parallelism
Reza Asad, Reza Babanezhad, Issam Laradji, Nicolas Le Roux, Sharan Vaswani
"Optimization for Machine Learning" workshop, NeurIPS, 2023. [pdf]

MSL: An Adaptive Momentem-based Stochastic Line-search Framework
Chen Fan, Sharan Vaswani , Christos Thrampoulidis, Mark Schmidt.
"Optimization for Machine Learning" workshop, NeurIPS, 2023. [pdf]

Target-based Surrogates for Stochastic Optimization
Jonathan Wilder Lavington^*, Sharan Vaswani^*, Reza Babanezhad, Mark Schmidt, Nicolas Le Roux, ICML, 2023. [pdf] [slides]
"Optimization for Machine Learning" workshop, NeurIPS 2022.

2022

Near-Optimal Sample Complexity Bounds for Constrained MDPs
Sharan Vaswani^*, Lin F. Yang^*, Csaba Szepesvari, NeurIPS, 2022. [pdf] [Csaba's talk at Simons]

Improved Policy Optimization for Online Imitation Learning
Jonathan Wilder Lavington, Sharan Vaswani, Mark Schmidt, CoLLAs, 2022. [pdf] [code]

Towards Painless Policy Optimization for Constrained MDPs
Arushi Jain^*, Sharan Vaswani^*, Reza Babanezhad, Csaba Szepesvari, Doina Precup, UAI, 2022. [pdf] [code]

Towards Noise-adaptive, Problem-adaptive (Accelerated) Stochastic Gradient Descent
Sharan Vaswani, Benjamin Dubois-Taine, Reza Babanezhad, ICML, 2022 (Oral Presentation). [pdf] [code] [slides]
"Optimization for Machine Learning" workshop, NeurIPS, 2021.

A general class of surrogate functions for stable and efficient reinforcement learning
Sharan Vaswani , Olivier Bachem, Simone Totaro, Robert Muller, Shivam Garg, Matthieu Geist, Marlos Machado, Pablo Samuel Castro, Nicolas Le Roux, AISTATS, 2022 (Best Paper Honorable Mention). [pdf] [code] [slides] [Talk at Alberta]
"Workshop on Reinforcement Learning Theory", ICML 2021.

SVRG meets AdaGrad: Painless Variance Reduction
Benjamin Dubois-Taine^*, Sharan Vaswani^*, Reza Babanezhad, Mark Schmidt, Simon Lacoste-Julien, , Machine Learning Journal 2022. [pdf] [code]

2021

Stochastic Polyak Step-size for SGD: An Adaptive Learning Rate for Fast Convergence
Nicolas Loizou, Sharan Vaswani , Issam Laradji, Simon Lacoste-Julien, AISTATS, 2021. [pdf] [code]
"Optimization for Machine Learning" workshop, NeurIPS 2020 (Spotlight) .

2020

Adaptive Gradient Methods Converge Faster with Over-Parameterization (but you should do a line-search)
Sharan Vaswani, Issam Laradji, Frederik Kunstner, Si Yi Meng, Mark Schmidt, Simon Lacoste-Julien, arXiv, 2020. [pdf] [code]
"Optimization for Machine Learning" workshop, NeurIPS 2020 (Spotlight) .

To Each Optimizer a Norm, To Each Norm its Generalization
Sharan Vaswani, Reza Babanezhad, Jose Gallego, Aaron Mishkin, Simon Lacoste-Julien, Nicolas Le Roux, arXiv, 2020. [pdf]
"Optimization for Machine Learning" workshop, NeurIPS 2020 (Spotlight) .

Old Dog Learns New Tricks: Randomized UCB for Bandit Problems
Sharan Vaswani, Abbas Mehrabian, Audrey Durand, Branislav Kveton, AISTATS 2020. [pdf] [code] [slides]

Fast and Furious Convergence: Stochastic Second Order Methods under Interpolation
Si Yi Meng^*, Sharan Vaswani^*, Issam Laradji, Mark Schmidt, Simon Lacoste-Julien, AISTATS 2020. [pdf] [code] [slides]
"Beyond First Order Methods in Machine Learning" workshop, NeurIPS 2019 (Spotlight) .

Combining Bayesian Optimization and Lipschitz Optimization
Mohamed Osama Ahmed, Sharan Vaswani, Mark Schmidt, Machine Learning Journal 2020. [pdf]

2019

Accelerating boosting via accelerated greedy coordinate descent
Xiaomeng Ju^*, Yifan Sun^*, Sharan Vaswani^*, Mark Schmidt. Optimization for Machine Learning workshop, NeurIPS 2019. [pdf] [code]

Painless Stochastic Gradient: Interpolation, Line-Search, and Convergence Rates
Sharan Vaswani, Aaron Mishkin, Issam Laradji, Mark Schmidt, Gauthier Gidel, Simon Lacoste-Julien, NeurIPS, 2019. [pdf] [code] [poster] [slides] [video]

Garbage In, Reward Out: Bootstrapping Exploration in Multi-Armed Bandits
Branislav Kveton, Csaba Szepesvari, Sharan Vaswani, Zheng Wen, Mohammad Ghavamzadeh, Tor Lattimore, ICML, 2019. [pdf]

Fast and Faster Convergence of SGD for Over-Parameterized Models and an Accelerated Perceptron
Sharan Vaswani, Francis Bach, Mark Schmidt, AISTATS, 2019. [pdf] [poster]

2018

Structured Bandits and Applications
Sharan Vaswani, PhD thesis, University of British Columbia. [pdf] [slides]

New Insights into Bootstrapping for Bandits
Sharan Vaswani, Branislav Kveton, Zheng Wen, Anup Rao, Mark Schmidt, Yasin Abbasi-Yadkori, arXiv, 2018. [pdf]

2017

Online Influence Maximization under Independent Cascade Model with Semi-Bandit Feedback
Zheng Wen, Branislav Kveton, Michal Valko, Sharan Vaswani, NIPS, 2017. [pdf] [poster][article]

Model-Independent Online Learning for Influence Maximization
Sharan Vaswani, Branislav Kveton, Zheng Wen, Mohammad Ghavamzadeh, Laks.V.S.Lakshmanan, Mark Schmidt, ICML, 2017. [pdf] [slides] [poster] [code]

Horde of Bandits using Gaussian Markov Random Fields
Sharan Vaswani, Mark Schmidt, Laks.V.S.Lakshmanan, AISTATS, 2017. (Oral Presentation) [pdf] [poster] [slides]

2016

Adaptive Influence Maximization: Why commit when you can adapt ?
Sharan Vaswani, Laks.V.S.Lakshmanan, arXiv, 2016. [pdf]

2015

Influence Maximization with Bandits
Sharan Vaswani, Laks.V.S.Lakshmanan, Mark Schmidt, NIPS workshop on Networks in the Social and Information Sciences 2015, arXiv, 2015. [pdf]

Influence Maximization in Bandit and Adaptive Settings
Sharan Vaswani, MSc thesis, University of British Columbia. [pdf]

2014

Modeling Non-Progressive Phenomena for Influence Propagation
Vincent Yun Lou, Smriti Bhagat, Laks V.S. Lakshmanan,Sharan Vaswani, ACM Conference on Online Social Networks(COSN) 2014. [pdf] [ Tech_report ]

Previous Research

Fast 3D Salient Region Detection in Medical Images using GPUs
Rahul Thota, Sharan Vaswani, Amit Kale, and Nagavijayalakshmi Vydyanathan. Machine Intelligence and Signal Processing. SpringerIndia, 2016. [pdf]

Performance Evaluation of Medical Imaging Algorithms on Intel MIC Platform
Jyotsna Khemka, Mrugesh Gajjar, Sharan Vaswani, Nagavijayalakshmi Vydyanathan, Rama Malladi, Vinutha V. 20th IEEE International Conference on High Performance Computing (HiPC) 2013. [pdf]

Fast 3D Structure Localization in Medical Volumes using CUDA-enabled GPUs
Sharan Vaswani, Rahul Thota, Nagavijayalakshmi Vydyanathan, Amit Kale. 2nd IEEE International Conference on Parallel Distributed and Grid Computing 2012, India. (Best Paper Award). [pdf]