I am interested in all theoretical aspects of Machine Learning.
Publications
A. Preprints
From Optimization to Sampling via Lyapunov Potentials
August Chen, Karthik Sridharan Link
Online Learning with Unknown Constraints Karthik Sridharan, Seung Won Wilson Yoo Link
Langevin dynamics: A unified perspective on optimization via Lyapunov potentials
August Chen, Ayush Sekhari, Karthik Sridharan Link
On Sequential Probability Assignment with Binary Alphabets and Large Classes of Experts Alexander Rakhlin, Karthik Sridharan [pdf]
Online Nonparametric Regression with General Loss Functions Alexander Rakhlin, Karthik Sridharan [Arxiv]
On Convex Optimization, Fat Shattering and Learning Nathan Srebro, Karthik Sridharan [pdf]
B. Conferences
Selective Sampling and Imitation Learning via Online Regression
Ayush Sekhari, Karthik Sridharan, Wen Sun, and Runzhe Wu
Neural Information Processing Systems (NeurIPS), 2023
Contextual Bandits and Imitation Learning via Preference-Based Active Queries
Ayush Sekhari, Karthik Sridharan, Wen Sun, and Runzhe Wu
Neural Information Processing Systems (NeurIPS), 2023
From Gradient Flow on Population Loss to Learning with Stochastic Gradient Descent
Chris De Sa, Satyen Kale, Jason D. Lee, Ayush Sekhari, Karthik Sridharan
Neural Information Processing Systems (NeurIPS), 2022
On the Complexity of Adversarial Decision Making (full oral)
Dylan J. Foster, Alexander Rakhlin, Ayush Sekhari, Karthik Sridharan
Neural Information Processing Systems (NeurIPS), 2022
Guarantees for Epsilon-Greedy Reinforcement Learning with Function Approximation
Christoph Dann, Yishay Mansour, Mehryar Mohri, Ayush Sekhari, Karthik Sridharan
International Conference on Machine Learning (ICML 2022) . Short version at RLDM 2022 - Reinforcement Learning and Decision Making conference.
SGD: The role of Implicit Regularization, Batch-size and Multiple Epochs
Satyen Kale, Ayush Sekhari, Karthik Sridharan
Neural Information Processing Systems (NeurIPS), 2021
Agnostic Reinforcement Learning with Low-Rank MDPs and Rich Observations
Christoph Dann, Yishay Mansour, Mehryar Mohri, Ayush Sekhari, Karthik Sridharan
Neural Information Processing Systems (NeurIPS), 2021
Online learning with dynamics: A minimax perspective
Kush Bhatia, Karthik Sridharan
Neural Information Processing Systems (NeurIPS), 2020
Reinforcement Learning with Feedback Graphs
Christoph Dann, Yishay Mansour, Mehryar Mohri, Ayush Sekhari, Karthik Sridharan
NeurIPS 2020. Short version at ICML 2020 Theoretical Foundations of RL workshop.
Second-Order Information in Non-Convex Stochastic Optimization: Power and Limitations
Yossi Arjevani, Yair Carmon, John C Duchi, Dylan J Foster, Ayush Sekhari, Karthik Sridharan
COLT 2020.
Hypothesis Set Stability and Generalization
Dylan J. Foster, Spencer Greenberg, Satyen Kale, Haipeng Luo, Mehryar Mohri, Karthik Sridharan
NeurIPS 2019 [arxiv]
The Complexity of Making the Gradient Small in Stochastic Convex Optimization (best student paper award) Dylan J. Foster, Ayush Sekhari, Ohad Shamir, Nathan Srebro, Karthik Sridharan, Blake Woodworth
Conference on Learning Theory (COLT 2019) [arxiv]
Distributed Learning with Sublinear Communication (full oral presentation) Jayadev Acharya, Christopher De Sa, Dylan J. Foster, Karthik Sridharan
International Conference on Machine Learning (ICML 2019) [arxiv]
Training Well-Generalizing Classifiers for Fairness Metrics and Other Data-Dependent Constraints
Andrew Cotter, Maya Gupta, Heinrich Jiang, Nathan Srebro, Karthik Sridharan, Serena Wang, Blake Woodworth, Seungil You
International Conference on Machine Learning (ICML 2019) [arxiv]
Two-Player Games for Efficient Non-Convex Constrained Optimization (best paper award)
Andrew Cotter, Heinrich Jiang, Karthik Sridharan
Algorithmic Learning Theory (ALT 2019)
Uniform Convergence of Gradients for Non-Convex Learning and Optimization
Dylan Foster, Ayush Sekhari, Karthik Sridharan
NeurIPS 2018
Online Learning: Sufficient Statistics and the Burkholder Method
Dylan Foster, Alexander Rakhlin, Karthik Sridharan
COLT 2018
Logistic Regression: The Importance of Being Improper (best student paper award)
Dylan Foster, Satyen Kale, Haipeng Luo, Mehryar Mohri, Karthik Sridharan
COLT 2018
Small-loss bounds for online learning with partial information
Thodoris Lykouris, Karthik Sridharan, Eva Tardos
COLT 2018
Inference in Sparse Graphs with Pairwise Measurements and Side Information
Dylan Foster, Daniel Reichman, Karthik Sridharan
AISTATS 2018, [Arxiv]
Parameter-Free Online Learning via Model Selection
Dylan Foster, Satyen Kale, Mehryar Mohri, Karthik Sridharan
NIPS 2017
ZigZag: A new approach to adaptive online learning Dylan Foster, Alexander Rakhlin, Karthik Sridharan
Conference on Learning Theory (COLT) 2017 [ArXiv]
On Equivalence of Martingale Tail Bounds and Deterministic Regret Inequalities Alexander Rakhlin, Karthik Sridharan
Conference on Learning Theory (COLT) 2017 [ArXiv]
Efficient Multiclass Prediction on Graphs via Surrogate Losses Alexander Rakhlin, Karthik Sridharan
AISTATS 2017 [PDF]
Learning in Games: Robustness of Fast Convergence
Dylan Foster, Zhiyuan Li, Thodoris Lykouris, Karthik Sridharan, Eva Tardos
NIPS 2016
Exploiting the Structure: Stochastic Gradient Methods Using Raw Clusters Zeyuan Allen-Zhu*, Yang Yuan*, Karthik Sridharan
NIPS 2016 (* - main contributors)
BISTRO: An Efficient Relaxation-Based Method for Contextual Bandits Alexander Rakhlin, Karthik Sridharan
ICML 2016, [ArXiv]
Hierarchies of Relaxations for Online Prediction Problems with Evolving Constraints Alexander Rakhlin, Karthik Sridharan
COLT 2015, [arxiv]
Learning with Square Loss: Localization through Offset Rademacher Complexity Tengyuan Liang, Alexander Rakhlin, Karthik Sridharan
COLT 2015, [arxiv]
Online Optimization : Competing with Dynamic Comparators Ali Jadbabaie, Alexander Rakhlin, Shahin Shshrampour, Karthik Sridharan
AISTATS 2015, [pdf]
Online Nonparametric Regression Alexander Rakhlin, Karthik Sridharan
COLT 2014 [pdf]
On Semi-Probabilistic Universal Prediction Alexander Rakhlin, Karthik Sridharan
Proceedings of IEEE Information Theory Workshop, 2013. Invited paper [pdf]
Optimization, Learning, and Games with Predictable Sequences Alexander Rakhlin, Karthik Sridharan
NIPS 2013 [pdf]
Competing With Strategies Wei Han, Alexander Rakhlin, Karthik Sridharan
COLT 2013 [pdf]
Online Learning with Predictable Sequences Alexander Rakhlin, Karthik Sridharan COLT 2013
[pdf] , [Arxiv version]
Localization and Adaptation in Online Learning (full oral presentation) Alexander Rakhlin, Ohad Shamir, Karthik Sridharan AISTATS 2013
Relax and Randomize: From Value to Algorithms (full oral presentation) Alexander Rakhlin, Ohad Shamir, Karthik Sridharan NIPS 2012 [pdf]
Making Stochastic Gradient Descent Optimal for Strongly Convex Problems Alexander Rakhlin, Ohad Shamir, Karthik Sridharan
ICML 2012 [Arxiv Version]
Minimizing The Misclassification Error Rate Using a Surrogate Convex Loss Shai Ben-David, David Loker, Nathan Srebro, Karthik Sridharan
ICML 2012 [pdf]
On the Universality of Online Mirror Descent Nathan Srebro, Karthik Sridharan, Ambuj Tewari
NIPS 2011 [Arxiv Version]
Better Mini-Batch Algorithms via Accelerated Gradient Methods Andrew Cotter, Ohad Shamir , Nathan Srebro, Karthik Sridharan
NIPS 2011 [Arxiv Version]
Online Learning: Stochastic and Constrained Adversaries Alexander Rakhlin, Karthik Sridharan, Ambuj Tewari
NIPS 2011 [pdf][Arxiv Version]
Online Learning: Beyond Regret (best paper award) Alexander Rakhlin, Karthik Sridharan, Ambuj Tewari
COLT 2011, [pdf][Arxiv Version]
Complexity-Based Approach to Calibration with Checking Rules Dean Foster, Alexander Rakhlin, Karthik Sridharan, Ambuj Tewari
COLT 2011, [pdf]
Online Learning: Random Averages, Combinatorial Parameters and Learnability (full oral presentation) Alexander Rakhlin, Karthik Sridharan, Ambuj Tewari NIPS 2010 [pdf][Arxiv Version]
Smoothness, Low-Noise and Fast Rates Nathan Srebro, Karthik Sridharan, Ambuj Tewari NIPS 2010 [pdf][Arxiv version]
Robust Selective Sampling from Single and Multiple Teachers Ofer Dekel, Claudio Gentile, Karthik Sridharan COLT 2010 [pdf]
Convex Games in Banach Spaces Karthik Sridharan, Ambuj Tewari COLT 2010 [pdf]
Learning Kernel-Based Halfspaces with the Zero-One Loss (best paper award) Shai Shalev-Shwartz, Ohad Shamir, Karthik Sridharan COLT 2010 [pdf], A shorter version presented at the best paper track IJCAI 2011 [pdf]
On the Complexity of Linear Prediction: Risk Bounds, Margin Bounds and Regularization Sham Kakade, Karthik Sridharan, Ambuj Tewari NIPS 2008 [pdf]
Fast Rates for Regularized Objectives Shai Shalev-Shwartz, Nathan Srebro, Karthik Sridharan NIPS 2008 [pdf]
Information Theoretic Framework for Multi-view Learning Karthik Sridharan, Sham M. Kakade 21st Annual Conference on Learning Theory (COLT 2008) [pdf]
Competitive Mixtures of Simple Neurons Karthik Sridharan, Matthew J Beal, Venu Govindaraju ICPR'06 [pdf]
Identifying handwritten text in mixed documents Faisal Farooq, Karthik Sridharan, Venu Govindaraju ICPR'06
Classification of Machine Print and Handwritten Arabic Documents Karthik Sridharan, Faisal Farooq, Venu Govindaraju (SDIUT 2005, pp. 89-94.)
A Sampling Based Approach to Facial Feature Extraction IEEE link Karthik Sridharan, Venu Govindaraju (IEEE AUTOID 2005. Best Paper Award - Second Prize, pp.51-56)
A Probabilistic Approach to Semantic Face Retrievalspringer link Karthik Sridharan, Sankalp Nayak, Sharat Chikkerur, Venu Govindaraju (AVBPA 2005, pp.977-986.)
A Dynamic Migration Model for Self-adaptive Genetic Algorithmsspringer link K.G. Srinivasa, Karthik Sridharan, P. Deepa Shenoy, Venugopal K.R., L.M. Patnaik (Proceedings of 6th International Conference on Intelligent Data Engineering
and Automated Learning (IDEAL 05), Springer Verlag, LNCS, July 6th – 9th 2005, Brisbane,
Australia, pp. 555-562.)
An Effective Content-Based Image Retrieval System Using STI Features and Relevance Feedback K.G. Srinivasa, Karthik Sridharan, P. Deepa Shenoy, Venugopal K.R., L.M. Patnaik (KBCS-2004, Fifth International Conference On Knowledge Based Computer Systems, Hyderabad, India, December 19-22, 2004, pp. 290 - 301.)
EASOM: An Efficient Soft Computing Method for Predicting the Share ValuesACTA press link K.G. Srinivasa, Karthik Sridharan, P. Deepa Shenoy, Venugopal K.R., L.M. Patnaik (Proceedings of IASTED International Conference on Artificial Intelligence and Applications (AIA 2004), ISSN: 1027-2666, Austria, Innsburg, Feb 16 - 18, 2004, pp. 264-269.)
C. Journals
Optimization with Non-Differentiable Constraints with Applications to Fairness, Recall, Churn, and Other Goals
Andrew Cotter, Heinrich Jiang, Serena Wang, Taman Narayan, Maya Gupta, Seungil You, Karthik Sridharan
Journal of Machine Learning Research, (to appear) 2019
Empirical Entropy, Minimax Regret and Minimax Risk Alexander Rakhlin, Karthik Sridharan, Alexandre Tsybakov
Bernoulli Journal, 2014 (to appear) [pdf]
Online Learning via Sequential Complexities
Alexander Rakhlin, Karthik Sridharan, Ambuj Tewari
Journal of Machine Learning Research, 2014
Sequential Complexities and Uniform Martingale Laws of Large
Numbers Alexander Rakhlin, Karthik Sridharan, Ambuj Tewari
Probability Theory and Related Fields, 2014, to appear [pdf]
Selective Sampling and Active Learning from Single and Multiple Teachers Ofer Dekel, Claudio Gentile, Karthik Sridharan Journal of Machine Learning Research, 2012 [pdf]
Learning Kernel Based Halfspaces with the 0-1 Loss Shai Shalev-Shwartz, Ohad Shamir, Karthik Sridharan SIAM Journal on Computing, 40(6):1623-1646, 2011 [pdf]
A Neural Network based CBIR System using STI Features and Relevance Feedback K.G. Srinivasa, Karthik Sridharan, P. Deepa Shenoy, Venugopal K.R., L.M. Patnaik International Journal on Intelligent Data Analysis, Volume 10, Number 2, 2006, IOS Press.
D. Thesis
Doctoral Thesis : Learning from an Optimization Viewpoint Karthik Sridharan Advisor : Nati Srebro Thesis Commitee : David McAllester, Arkadi Nemirovski, Alexander Razborov, Nathan Srebro
Toyota Technological Institute at Chicago [pdf]
A Tutorial on Online Supervised Learning with Applications to Node Classification in Social Networks
Alexander Rakhlin, Karthik Sridharan [Arxiv]
Note on Refined Dudley Integral Covering Number Bound Nathan Srebro, Karthik Sridharan [pdf]
A Gentle Introduction to Concentration Inequalities Karthik Sridharan (Theorems and proofs of a few concentration inequalities) - [pdf] [PS]
Fast Convergence Rates for Excess Regularized Risk with Application to SVM Karthik Sridharan [pdf]
F. Books and Book Chapters
Statistical Learning Theory and Sequential Prediction Alexander Rakhlin, Karthik Sridharan
Lecture Notes (Early book draft) [pdf]
On Martingale Extensions of Vapnik-Chervonenkis Theory with Applications to Online Learning Alexander Rakhlin, Karthik Sridharan
Chapter 15, pp 197-215,: "Measures of Complexity", Festschrift for Alexey Chervonenkis. the book , Chapter [pdf]
G. Selected Current Projects
Plug & Play Machine Learning
Online Learning on Graphs and other interconnected instances