My interests are in the intersection of algorithms, statistics, optimization, and machine learning. Improves the stochas-tic convex optimization problem in parallel and DP setting. SHUFE, Oct. 2022 - Algorithm Seminar, Google Research, Oct. 2022 - Young Researcher Workshop, Cornell ORIE, Apr. SODA 2023: 5068-5089. [pdf] Many of my results use fast matrix multiplication Huang Engineering Center Oral Presentation for Misspecification in Prediction Problems and Robustness via Improper Learning. with Yair Carmon, Kevin Tian and Aaron Sidford Journal of Machine Learning Research, 2017 (arXiv). Multicalibrated Partitions for Importance Weights Parikshit Gopalan, Omer Reingold, Vatsal Sharan, Udi Wieder ALT, 2022 arXiv . In particular, this work presents a sharp analysis of: (1) mini-batching, a method of averaging many . ", "General variance reduction framework for solving saddle-point problems & Improved runtimes for matrix games. Google Scholar; Probability on trees and . Algorithms Optimization and Numerical Analysis. He received his PhD from the Electrical Engineering and Computer Science Department at the Massachusetts Institute of Technology, where he was advised by Jonathan Kelner. Authors: Michael B. Cohen, Jonathan Kelner, Rasmus Kyng, John Peebles, Richard Peng, Anup B. Rao, Aaron Sidford Download PDF Abstract: We show how to solve directed Laplacian systems in nearly-linear time. [pdf] [poster] I am broadly interested in optimization problems, sometimes in the intersection with machine learning theory and graph applications. About Me. ", "A low-bias low-cost estimator of subproblem solution suffices for acceleration! with Aaron Sidford Some I am still actively improving and all of them I am happy to continue polishing. In Sidford's dissertation, Iterative Methods, Combinatorial . I am a fifth year Ph.D. student in Computer Science at Stanford University co-advised by Gregory Valiant and John Duchi. IEEE, 147-156. Verified email at stanford.edu - Homepage. 2023. . NeurIPS Smooth Games Optimization and Machine Learning Workshop, 2019, Variance Reduction for Matrix Games Yair Carmon. Department of Electrical Engineering, Stanford University, 94305, Stanford, CA, USA Associate Professor of . With Yosheb Getachew, Yujia Jin, Aaron Sidford, and Kevin Tian (2023). The Complexity of Infinite-Horizon General-Sum Stochastic Games, With Yujia Jin, Vidya Muthukumar, Aaron Sidford, To appear in Innovations in Theoretical Computer Science (ITCS 2023) (arXiv), Optimal and Adaptive Monteiro-Svaiter Acceleration, With Yair Carmon, Danielle Hausler, Arun Jambulapati, and Yujia Jin, To appear in Advances in Neural Information Processing Systems (NeurIPS 2022) (arXiv), On the Efficient Implementation of High Accuracy Optimality of Profile Maximum Likelihood, With Moses Charikar, Zhihao Jiang, and Kirankumar Shiragur, Improved Lower Bounds for Submodular Function Minimization, With Deeparnab Chakrabarty, Andrei Graur, and Haotian Jiang, In Symposium on Foundations of Computer Science (FOCS 2022) (arXiv), RECAPP: Crafting a More Efficient Catalyst for Convex Optimization, With Yair Carmon, Arun Jambulapati, and Yujia Jin, International Conference on Machine Learning (ICML 2022) (arXiv), Efficient Convex Optimization Requires Superlinear Memory, With Annie Marsden, Vatsal Sharan, and Gregory Valiant, Conference on Learning Theory (COLT 2022), Sharper Rates for Separable Minimax and Finite Sum Optimization via Primal-Dual Extragradient Method, Conference on Learning Theory (COLT 2022) (arXiv), Big-Step-Little-Step: Efficient Gradient Methods for Objectives with Multiple Scales, With Jonathan A. Kelner, Annie Marsden, Vatsal Sharan, Gregory Valiant, and Honglin Yuan, Regularized Box-Simplex Games and Dynamic Decremental Bipartite Matching, With Arun Jambulapati, Yujia Jin, and Kevin Tian, International Colloquium on Automata, Languages and Programming (ICALP 2022) (arXiv), Fully-Dynamic Graph Sparsifiers Against an Adaptive Adversary, With Aaron Bernstein, Jan van den Brand, Maximilian Probst, Danupon Nanongkai, Thatchaphol Saranurak, and He Sun, Faster Maxflow via Improved Dynamic Spectral Vertex Sparsifiers, With Jan van den Brand, Yu Gao, Arun Jambulapati, Yin Tat Lee, Yang P. Liu, and Richard Peng, In Symposium on Theory of Computing (STOC 2022) (arXiv), Semi-Streaming Bipartite Matching in Fewer Passes and Optimal Space, With Sepehr Assadi, Arun Jambulapati, Yujia Jin, and Kevin Tian, In Symposium on Discrete Algorithms (SODA 2022) (arXiv), Algorithmic trade-offs for girth approximation in undirected graphs, With Avi Kadria, Liam Roditty, Virginia Vassilevska Williams, and Uri Zwick, In Symposium on Discrete Algorithms (SODA 2022), Computing Lewis Weights to High Precision, With Maryam Fazel, Yin Tat Lee, and Swati Padmanabhan, With Hilal Asi, Yair Carmon, Arun Jambulapati, and Yujia Jin, In Advances in Neural Information Processing Systems (NeurIPS 2021) (arXiv), Thinking Inside the Ball: Near-Optimal Minimization of the Maximal Loss, In Conference on Learning Theory (COLT 2021) (arXiv), The Bethe and Sinkhorn Permanents of Low Rank Matrices and Implications for Profile Maximum Likelihood, With Nima Anari, Moses Charikar, and Kirankumar Shiragur, Towards Tight Bounds on the Sample Complexity of Average-reward MDPs, In International Conference on Machine Learning (ICML 2021) (arXiv), Minimum cost flows, MDPs, and 1-regression in nearly linear time for dense instances, With Jan van den Brand, Yin Tat Lee, Yang P. Liu, Thatchaphol Saranurak, and Zhao Song, Di Wang, In Symposium on Theory of Computing (STOC 2021) (arXiv), Ultrasparse Ultrasparsifiers and Faster Laplacian System Solvers, In Symposium on Discrete Algorithms (SODA 2021) (arXiv), Relative Lipschitzness in Extragradient Methods and a Direct Recipe for Acceleration, In Innovations in Theoretical Computer Science (ITCS 2021) (arXiv), Acceleration with a Ball Optimization Oracle, With Yair Carmon, Arun Jambulapati, Qijia Jiang, Yujia Jin, Yin Tat Lee, and Kevin Tian, In Conference on Neural Information Processing Systems (NeurIPS 2020), Instance Based Approximations to Profile Maximum Likelihood, In Conference on Neural Information Processing Systems (NeurIPS 2020) (arXiv), Large-Scale Methods for Distributionally Robust Optimization, With Daniel Levy*, Yair Carmon*, and John C. Duch (* denotes equal contribution), High-precision Estimation of Random Walks in Small Space, With AmirMahdi Ahmadinejad, Jonathan A. Kelner, Jack Murtagh, John Peebles, and Salil P. Vadhan, In Symposium on Foundations of Computer Science (FOCS 2020) (arXiv), Bipartite Matching in Nearly-linear Time on Moderately Dense Graphs, With Jan van den Brand, Yin Tat Lee, Danupon Nanongkai, Richard Peng, Thatchaphol Saranurak, Zhao Song, and Di Wang, In Symposium on Foundations of Computer Science (FOCS 2020), With Yair Carmon, Yujia Jin, and Kevin Tian, Unit Capacity Maxflow in Almost $O(m^{4/3})$ Time, Invited to the special issue (arXiv before merge)), Solving Discounted Stochastic Two-Player Games with Near-Optimal Time and Sample Complexity, In International Conference on Artificial Intelligence and Statistics (AISTATS 2020) (arXiv), Efficiently Solving MDPs with Stochastic Mirror Descent, In International Conference on Machine Learning (ICML 2020) (arXiv), Near-Optimal Methods for Minimizing Star-Convex Functions and Beyond, With Oliver Hinder and Nimit Sharad Sohoni, In Conference on Learning Theory (COLT 2020) (arXiv), Solving Tall Dense Linear Programs in Nearly Linear Time, With Jan van den Brand, Yin Tat Lee, and Zhao Song, In Symposium on Theory of Computing (STOC 2020). [pdf] [poster] 2013. pdf, Fourier Transformation at a Representation, Annie Marsden. Stanford, CA 94305 Internatioonal Conference of Machine Learning (ICML), 2022, Semi-Streaming Bipartite Matching in Fewer Passes and Optimal Space University, Research Institute for Interdisciplinary Sciences (RIIS) at . Here are some lecture notes that I have written over the years. with Yair Carmon, Arun Jambulapati, Qijia Jiang, Yin Tat Lee, Aaron Sidford and Kevin Tian Faster energy maximization for faster maximum flow. Two months later, he was found lying in a creek, dead from . I am particularly interested in work at the intersection of continuous optimization, graph theory, numerical linear algebra, and data structures. with Hilal Asi, Yair Carmon, Arun Jambulapati and Aaron Sidford (, In Symposium on Foundations of Computer Science (FOCS 2015) (, In Conference on Learning Theory (COLT 2015) (, In International Conference on Machine Learning (ICML 2015) (, In Innovations in Theoretical Computer Science (ITCS 2015) (, In Symposium on Fondations of Computer Science (FOCS 2013) (, In Symposium on the Theory of Computing (STOC 2013) (, Book chapter in Building Bridges II: Mathematics of Laszlo Lovasz, 2020 (, Journal of Machine Learning Research, 2017 (. Towards this goal, some fundamental questions need to be solved, such as how can machines learn models of their environments that are useful for performing tasks . Student Intranet. arXiv | conference pdf (alphabetical authorship) Jonathan Kelner, Annie Marsden, Vatsal Sharan, Aaron Sidford, Gregory Valiant, Honglin Yuan, Big-Step-Little-Step: Gradient Methods for Objectives with . Daniel Spielman Professor of Computer Science, Yale University Verified email at yale.edu. I completed my PhD at /N 3 Thesis, 2016. pdf. BayLearn, 2021, On the Sample Complexity of Average-reward MDPs With Cameron Musco and Christopher Musco. in Chemistry at the University of Chicago. Given a linear program with n variables, m > n constraints, and bit complexity L, our algorithm runs in (sqrt(n) L) iterations each consisting of solving (1) linear systems and additional nearly linear time computation. of practical importance. Aaron Sidford's 143 research works with 2,861 citations and 1,915 reads, including: Singular Value Approximation and Reducing Directed to Undirected Graph Sparsification I am fortunate to be advised by Aaron Sidford . [pdf] 2021. Yujia Jin. Aaron Sidford is an assistant professor in the departments of Management Science and Engineering and Computer Science at Stanford University. July 2015. pdf, Szemerdi Regularity Lemma and Arthimetic Progressions, Annie Marsden. I am a fifth-and-final-year PhD student in the Department of Management Science and Engineering at Stanford in with Hilal Asi, Yair Carmon, Arun Jambulapati and Aaron Sidford The paper, Efficient Convex Optimization Requires Superlinear Memory, was co-authored with Stanford professor Gregory Valiant as well as current Stanford student Annie Marsden and alumnus Vatsal Sharan. In Symposium on Theory of Computing (STOC 2020) (arXiv), Constant Girth Approximation for Directed Graphs in Subquadratic Time, With Shiri Chechik, Yang P. Liu, and Omer Rotem, Leverage Score Sampling for Faster Accelerated Regression and ERM, With Naman Agarwal, Sham Kakade, Rahul Kidambi, Yin Tat Lee, and Praneeth Netrapalli, In International Conference on Algorithmic Learning Theory (ALT 2020) (arXiv), Near-optimal Approximate Discrete and Continuous Submodular Function Minimization, In Symposium on Discrete Algorithms (SODA 2020) (arXiv), Fast and Space Efficient Spectral Sparsification in Dynamic Streams, With Michael Kapralov, Aida Mousavifar, Cameron Musco, Christopher Musco, Navid Nouri, and Jakab Tardos, In Conference on Neural Information Processing Systems (NeurIPS 2019), Complexity of Highly Parallel Non-Smooth Convex Optimization, With Sbastien Bubeck, Qijia Jiang, Yin Tat Lee, and Yuanzhi Li, Principal Component Projection and Regression in Nearly Linear Time through Asymmetric SVRG, A Direct (1/) Iteration Parallel Algorithm for Optimal Transport, In Conference on Neural Information Processing Systems (NeurIPS 2019) (arXiv), A General Framework for Efficient Symmetric Property Estimation, With Moses Charikar and Kirankumar Shiragur, Parallel Reachability in Almost Linear Work and Square Root Depth, In Symposium on Foundations of Computer Science (FOCS 2019) (arXiv), With Deeparnab Chakrabarty, Yin Tat Lee, Sahil Singla, and Sam Chiu-wai Wong, Deterministic Approximation of Random Walks in Small Space, With Jack Murtagh, Omer Reingold, and Salil P. Vadhan, In International Workshop on Randomization and Computation (RANDOM 2019), A Rank-1 Sketch for Matrix Multiplicative Weights, With Yair Carmon, John C. Duchi, and Kevin Tian, In Conference on Learning Theory (COLT 2019) (arXiv), Near-optimal method for highly smooth convex optimization, Efficient profile maximum likelihood for universal symmetric property estimation, In Symposium on Theory of Computing (STOC 2019) (arXiv), Memory-sample tradeoffs for linear regression with small error, Perron-Frobenius Theory in Nearly Linear Time: Positive Eigenvectors, M-matrices, Graph Kernels, and Other Applications, With AmirMahdi Ahmadinejad, Arun Jambulapati, and Amin Saberi, In Symposium on Discrete Algorithms (SODA 2019) (arXiv), Exploiting Numerical Sparsity for Efficient Learning: Faster Eigenvector Computation and Regression, In Conference on Neural Information Processing Systems (NeurIPS 2018) (arXiv), Near-Optimal Time and Sample Complexities for Solving Discounted Markov Decision Process with a Generative Model, With Mengdi Wang, Xian Wu, Lin F. Yang, and Yinyu Ye, Coordinate Methods for Accelerating Regression and Faster Approximate Maximum Flow, In Symposium on Foundations of Computer Science (FOCS 2018), Solving Directed Laplacian Systems in Nearly-Linear Time through Sparse LU Factorizations, With Michael B. Cohen, Jonathan A. Kelner, Rasmus Kyng, John Peebles, Richard Peng, and Anup B. Rao, In Symposium on Foundations of Computer Science (FOCS 2018) (arXiv), Efficient Convex Optimization with Membership Oracles, In Conference on Learning Theory (COLT 2018) (arXiv), Accelerating Stochastic Gradient Descent for Least Squares Regression, With Prateek Jain, Sham M. Kakade, Rahul Kidambi, and Praneeth Netrapalli, Approximating Cycles in Directed Graphs: Fast Algorithms for Girth and Roundtrip Spanners. CV (last updated 01-2022): PDF Contact. Research interests : Data streams, machine learning, numerical linear algebra, sketching, and sparse recovery.. This work presents an accelerated gradient method for nonconvex optimization problems with Lipschitz continuous first and second derivatives that is Hessian free, i.e., it only requires gradient computations, and is therefore suitable for large-scale applications. [pdf] [talk] [poster] Articles 1-20. ", "Collection of new upper and lower sample complexity bounds for solving average-reward MDPs. D Garber, E Hazan, C Jin, SM Kakade, C Musco, P Netrapalli, A Sidford. missouri noodling association president cnn. ", "Collection of variance-reduced / coordinate methods for solving matrix games, with simplex or Euclidean ball domains. I often do not respond to emails about applications. Secured intranet portal for faculty, staff and students. We establish lower bounds on the complexity of finding $$-stationary points of smooth, non-convex high-dimensional functions using first-order methods. COLT, 2022. Fall'22 8803 - Dynamic Algebraic Algorithms, small tool to obtain upper bounds of such algebraic algorithms. Discrete Mathematics and Algorithms: An Introduction to Combinatorial Optimization: I used these notes to accompany the course Discrete Mathematics and Algorithms. by Aaron Sidford. O! Google Scholar, The Complexity of Infinite-Horizon General-Sum Stochastic Games, The Complexity of Optimizing Single and Multi-player Games, A Near-Optimal Method for Minimizing the Maximum of N Convex Loss Functions, On the Sample Complexity for Average-reward Markov Decision Processes, Stochastic Methods for Matrix Games and its Applications, Acceleration with a Ball Optimization Oracle, Principal Component Projection and Regression in Nearly Linear Time through Asymmetric SVRG, The Complexity of Infinite-Horizon General-Sum Stochastic Games Some I am still actively improving and all of them I am happy to continue polishing. Research Institute for Interdisciplinary Sciences (RIIS) at CS265/CME309: Randomized Algorithms and Probabilistic Analysis, Fall 2019. with Yair Carmon, Arun Jambulapati and Aaron Sidford We are excited to have Professor Sidford join the Management Science & Engineering faculty starting Fall 2016. From 2016 to 2018, I also worked in Neural Information Processing Systems (NeurIPS), 2021, Thinking Inside the Ball: Near-Optimal Minimization of the Maximal Loss theory and graph applications. International Conference on Machine Learning (ICML), 2020, Principal Component Projection and Regression in Nearly Linear Time through Asymmetric SVRG Congratulations to Prof. Aaron Sidford for receiving the Best Paper Award at the 2022 Conference on Learning Theory (COLT 2022)! which is why I created a ", "How many \(\epsilon\)-length segments do you need to look at for finding an \(\epsilon\)-optimal minimizer of convex function on a line? The authors of most papers are ordered alphabetically. Yang P. Liu, Aaron Sidford, Department of Mathematics 2016. Np%p `a!2D4! I am a fifth-and-final-year PhD student in the Department of Management Science and Engineering at Stanford in the Operations Research group. With Jack Murtagh, Omer Reingold, and Salil P. Vadhan. pdf, Sequential Matrix Completion. Efficient accelerated coordinate descent methods and faster algorithms for solving linear systems. 172 Gates Computer Science Building 353 Jane Stanford Way Stanford University Congratulations to Prof. Aaron Sidford for receiving the Best Paper Award at the 2022 Conference on Learning Theory ( COLT 2022 )! This work characterizes the benefits of averaging techniques widely used in conjunction with stochastic gradient descent (SGD). 2017. We make safe shipping arrangements for your convenience from Baton Rouge, Louisiana. I maintain a mailing list for my graduate students and the broader Stanford community that it is interested in the work of my research group. Semantic parsing on Freebase from question-answer pairs. stream Slides from my talk at ITCS. %PDF-1.4 Li Chen, Rasmus Kyng, Yang P. Liu, Richard Peng, Maximilian Probst Gutenberg, Sushant Sachdeva, Online Edge Coloring via Tree Recurrences and Correlation Decay, STOC 2022 Roy Frostig, Sida Wang, Percy Liang, Chris Manning. with Yair Carmon, Danielle Hausler, Arun Jambulapati and Aaron Sidford I am currently a third-year graduate student in EECS at MIT working under the wonderful supervision of Ankur Moitra. Conference of Learning Theory (COLT), 2022, RECAPP: Crafting a More Efficient Catalyst for Convex Optimization SODA 2023: 4667-4767. Annie Marsden, Vatsal Sharan, Aaron Sidford, Gregory Valiant, Efficient Convex Optimization Requires Superlinear Memory. [pdf] with Yair Carmon, Aaron Sidford and Kevin Tian Full CV is available here. I am an assistant professor in the department of Management Science and Engineering and the department of Computer Science at Stanford University.