SS
Suvrit Sra
cs.LGmath.OCstat.MLmath.FAcs.AImath.CAmath.PRcs.CVmath.COcs.DS
On Valency
published · living versionsW_cmwskyw6·v1 · currentpublished
W_pr5z5huy·v1published
Preprints & journals
146 papers in the corpus · 2008–2026The -Bures-Wasserstein distance via block matrices (with application to -Bures-Wasserstein barycenters)2610.15889v1 · Suvrit Sra2026 · 0 citationsarXiv
A Proof of McLeod's 1959 Conjecture on Complete Homogeneous Symmetric Ratios2610.07302v1 · Suvrit Sra2026 · 0 citationsarXiv
A Koteljanskii inequality for permanents2609.39979v1 · Suvrit Sra2026 · 0 citationsarXiv
Tighter bounds on Komlós discrepancy: existence and algorithmic results2609.27172v2 · Emrullah Akbas, Suvrit Sra2026 · 0 citationsarXiv
Kadison--Singer partitions and Bilu--Linial graph signings in polynomial time2609.23855v1 · Ali Jadbabaie, Amin Saberi, Suvrit Sra2026 · 0 citationsarXiv
Boolean Small-Ball Inequalities for Discrepancy Theory2609.20785v1 · Emrullah Akbas, Suvrit Sra2026 · 0 citationsarXiv
Percolation Dynamics in Optimization: Variance Cascades and Nested Symmetry2609.02373v2 · Sai Niranjan Ramachandran, Suvrit Sra2026 · 0 citationsarXiv
GPT, the Counterexample Machine2608.29595v1 · Suvrit Sra2026 · 0 citationsarXiv
A Proof of the Matrix Spencer Conjecture2608.28816v1 · Emrullah Akbas, Suvrit Sra2026 · 0 citationsarXiv
An Algebraic Matrix Spencer Theorem2606.16005v1 · Emrullah Akbas, Suvrit Sra2026 · 0 citationsarXiv
Free Heavy-Tailed Lunch for Muon: A Theoretical Justification of Empirical Success2606.14560v1 · Florian Hubler, Thomas Pethick, Suvrit Sra2026 · 0 citationsarXiv
Active Inference as Context Acquisition for AI Agents2608.19202v1 · Sanchayan Dutta, Sai Niranjan Ramachandran, Suvrit Sra2026 · 0 citationsarXiv
Tight Generalization Bounds for Noiseless Inverse Optimization2605.08866v1 · Pouria Fatemi, Hoomaan Maskan, Suvrit Sra et al.2026 · 0 citationsarXiv
Trees to Flows and Back: Unifying Decision Trees and Diffusion Models2605.00414v2 · Sai Niranjan Ramachandran, Suvrit Sra2026 · 0 citationsarXiv
The Multi-Block DC Function Class: Theory, Algorithms, and Applications2604.17560v1 · Pouria Fatemi, Hoomaan Maskan, Alp Yurtsever et al.2026 · 0 citationsarXiv
Cost-Driven Representation Learning for Linear Quadratic Gaussian Control: Part II2603.07437v1 · Yi Tian, Kaiqing Zhang, Russ Tedrake et al.2026 · 0 citationsarXiv
Cross-fluctuation phase transitions reveal sampling dynamics in diffusion models2511.00124v1 · Sai Niranjan Ramachandran, Manish Krishan Lal, Suvrit Sra2025 · 0 citationsarXiv
Improved Rates for Stochastic Variance-Reduced Difference-of-Convex Algorithms2509.11657v1 · Anh Duc Nguyen, Alp Yurtsever, Suvrit Sra et al.2025 · 0 citationsarXiv
Linearly Convergent Algorithms for Nonsmooth Problems with Unknown Smooth Regions2507.19465v2 · Zhe Zhang, Suvrit Sra2025 · 0 citationsarXiv
A projection-based framework for gradient-free and parallel learning2506.05878v3 · Andreas Bergmeister, Manish Krishan Lal, Stefanie Jegelka et al.2025 · 0 citationsarXiv
Revisiting Frank-Wolfe for Structured Nonconvex Optimization2503.08921v2 · Hoomaan Maskan, Yikun Hou, Suvrit Sra et al.2025 · 0 citationsarXiv
Implicit Bias in Matrix Factorization and its Explicit Realization in a New Architecture2501.16322v2 · Yikun Hou, Suvrit Sra, Alp Yurtsever2025 · 0 citationsarXiv
Randomized Block Coordinate DC Programming2411.11664v3 · Hoomaan Maskan, Paniz Halvachi, Suvrit Sra et al.2024 · 0 citationsarXiv
Graph Transformers Dream of Electric Flow2410.16699v2 · Xiang Cheng, Lawrence Carin, Suvrit Sra2024 · 0 citationsarXiv
Toward generalizable learning of all (linear) first-order methods via memory augmented Transformers2410.07263v3 · Sanchayan Dutta (UC Davis), Suvrit Sra2024 · 0 citationsarXiv
First-Order Methods for Linearly Constrained Bilevel Optimization2406.12771v2 · Guy Kornowski, Swati Padmanabhan, Kai Wang et al.2024 · 2 citationsarXiv
Riemannian Bilevel Optimization2405.15816v1 · Sanchayan Dutta, Xiang Cheng, Suvrit Sra2024 · 0 citationsarXiv
Efficient Sampling on Riemannian Manifolds via Langevin MCMC2402.10357v1 · Xiang Cheng, Jingzhao Zhang, Suvrit Sra2024 · 3 citationsarXiv
Transformers Implement Functional Gradient Descent to Learn Non-Linear Functions In Context2312.06528v6 · Xiang Cheng, Yuxin Chen, Suvrit Sra2023 · 0 citationsarXiv
Linear attention is (maybe) all you need (to understand transformer optimization)2310.01082v2 · Kwangjun Ahn, Xiang Cheng, Minhak Song et al.2023 · 1 citationarXiv
Invex Programs: First Order Algorithms and Their Convergence2307.04456v1 · Adarsh Barik, Suvrit Sra, Jean Honorio2023 · 1 citationarXiv
Transformers learn to implement preconditioned gradient descent for in-context learning2306.00297v2 · Kwangjun Ahn, Xiang Cheng, Hadi Daneshmand et al.2023 · 11 citations37th Conference on Neural Information Processing Systems (NeurIPS 2023)
How to escape sharp minima with random perturbations2305.15659v3 · Kwangjun Ahn, Ali Jadbabaie, Suvrit Sra2023 · 0 citationsarXiv
The Crucial Role of Normalization in Sharpness-Aware Minimization2305.15287v2 · Yan Dai, Kwangjun Ahn, Suvrit Sra2023 · 1 citationarXiv
On the Training Instability of Shuffling SGD with Batch Normalization2302.12444v3 · David X. Wu, Chulhee Yun, Suvrit Sra2023 · 3 citationsarXiv
The Hornich-Hlawka functional inequality for functions with positive differences2301.08342v1 · Constantin P. Niculescu, Suvrit Sra2023 · 0 citationsarXiv
Cost-Driven Representation Learning for Linear Quadratic Gaussian Control: Part I2212.14511v4 · Yi Tian, Kaiqing Zhang, Russ Tedrake et al.2022 · 2 citationsarXiv
Computing Brascamp-Lieb Constants through the lens of Thompson Geometry2208.05013v3 · Melanie Weber, Suvrit Sra2022 · 1 citationarXiv
CCCP is Frank-Wolfe in disguise2206.12014v1 · Alp Yurtsever, Suvrit Sra2022 · 1 citationarXiv
On a class of geodesically convex optimization problems solved via Euclidean MM methods2206.11426v2 · Melanie Weber, Suvrit Sra2022 · 0 citationsarXiv
Theory and Algorithms for Diffusion Processes on Riemannian Manifolds2204.13665v3 · Xiang Cheng, Jingzhao Zhang, Suvrit Sra2022 · 1 citationarXiv
Understanding the unstable convergence of gradient descent2204.01050v2 · Kwangjun Ahn, Jingzhao Zhang, Suvrit Sra2022 · 9 citationsarXiv
Sign and Basis Invariant Networks for Spectral Graph Representation Learning2202.13013v4 · Derek Lim, Joshua Robinson, Lingxiao Zhao et al.2022 · 26 citationsarXiv
Sion's Minimax Theorem in Geodesic Metric Spaces and a Riemannian Extragradient Algorithm2202.06950v2 · Peiyuan Zhang, Jingzhao Zhang, Suvrit Sra2022 · 6 citationsarXiv
Time varying regression with hidden linear dynamics2112.14862v1 · Ali Jadbabaie, Horia Mania, Devavrat Shah et al.2021 · 2 citationsarXiv
Max-Margin Contrastive Learning2112.11450v1 · Anshul Shah, Suvrit Sra, Rama Chellappa et al.2021 · 43 citationsarXiv
Can contrastive learning avoid shortcut solutions?35546903 · Robinson, Joshua, Sun, Li, Yu, Ke et al.2024 · 48 citationsAdvances in neural information processing systems. 2021;34:4974-4986
Positive definite functions of noncommuting contractions, Hua-Bellman matrices, and a new distance metric2112.00056v2 · Suvrit Sra2021 · 0 citationsarXiv
Introducing Discrepancy Values of Matrices with Application to Bounding Norms of Commutators2111.11855v3 · Pourya Habib Zadeh, Suvrit Sra2021 · 0 citationsarXiv
Understanding Riemannian Acceleration via a Proximal Extragradient Framework2111.02763v2 · Jikai Jin, Suvrit Sra2021 · 1 citationarXiv
Career total: 350 works. 146 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.