Yeshwanth Cherapanamjeri

Yeshwanth Cherapanamjeri Postdoc CSAIL MIT I am a postdoc at MIT where I am mentored by Costis Daskalakis. I obtained my Ph.D at UC Berkeley where I was fortunate to be advised by Peter Bartlett. I'm broadly interested in algorithmic and statistical issues arising in machine learning. Before this, I'd spent two wonderful years as a research fellow in the Machine Learning & Optimization group at Microsoft Research, India, where I was fortunate to work with Prateek Jain, Praneeth Netrapalli and Nagarajan Natarajan on designing provable algorithms for non-convex optimization. Previously, I received a B. Tech in Computer Science from IIT Bombay where I worked on Entity Linking and Disambiguation with Prof. Ganesh Ramakrishnan and Prof. Soumen Chakrabarti for my undergraduate thesis. You can find a link to my resume here.

Publications

• Statistical Barriers to Affine-equivariant Estimation
  Zihao Chen, Yeshwanth Cherapanamjeri
  In Submission [ArXiv]

• An Investigation into the Effects of Pre-training Data Distributions for Pathology Report Classification
  Aliyah R. Hsu, Yeshwanth Cherapanamjeri, Briton Park, Tristan Naumann, Anobel Y. Odisho, Bin Yu
  In Submission [ArXiv]

• Optimal PAC Bounds Without Uniform Convergence
  Ishaq Aden-Ali, Yeshwanth Cherapanamjeri, Abhishek Shetty, Nikita Zhivotovskiy
  FOCS 2023 [ArXiv]
  Invited to SICOMP

• The One-Inclusion Graph Algorithm is not Always Optimal
  Ishaq Aden-Ali, Yeshwanth Cherapanamjeri, Abhishek Shetty, Nikita Zhivotovskiy
  COLT 2023 [ArXiv]

• Robust Algorithms on Adaptive Inputs from Bounded Adversaries
  Yeshwanth Cherapanamjeri, Sandeep Silwal, David P. Woodruff, Fred Zhang, Qiuyi Zhang, Samson Zhou
  ICLR 2023 [ArXiv]

• Optimal Algorithms for Linear Algebra in the Current Matrix Multiplication Time
  Yeshwanth Cherapanamjeri, Sandeep Silwal, David P. Woodruff, Samson Zhou
  SODA 2023 [ArXiv]

• What Makes A Good Fisherman? Linear Regression under Self-Selection Bias
  Yeshwanth Cherapanamjeri, Constantinos Daskalakis, Andrew Ilyas, Manolis Zampetakis
  STOC 2023 [ArXiv]

• Estimation of Standard Auction Models
  Yeshwanth Cherapanamjeri, Constantinos Daskalakis, Andrew Ilyas, Manolis Zampetakis
  Extended Abstract at EC 2022 [ArXiv]

• Uniform Approximations for Randomized Hadamard Transforms with Applications
  Yeshwanth Cherapanamjeri, Jelani Nelson
  STOC 2022 [ArXiv]

• Terminal Embeddings in Sublinear Time
  Yeshwanth Cherapanamjeri, Jelani Nelson
  FOCS 2021 [ArXiv]

• Adversarial Examples in Multi-Layer Random ReLU Networks
  Peter L. Bartlett, Sébastien Bubeck, Yeshwanth Cherapanamjeri
  NeurIPS 2021 [ArXiv]

• A single gradient step finds adversarial examples on random two-layers neural networks
  Sébastien Bubeck, Yeshwanth Cherapanamjeri, Gauthier Gidel, Rémi Tachet des Combes
  NeurIPS 2021 (Spotlight) [ArXiv]

• On Adaptive Distance Estimation
  Yeshwanth Cherapanamjeri, Jelani Nelson
  NeurIPS 2020 (Spotlight) [ArXiv]

• Optimal Robust Linear Regression in Nearly Linear Time
  Yeshwanth Cherapanamjeri, Efe Aras, Nilesh Tripuraneni, Michael I. Jordan, Nicolas Flammarion, Peter L. Bartlett
  In Submission [ArXiv]

• List Decodable Mean Estimation in Nearly Linear Time
  Yeshwanth Cherapanamjeri, Sidhanth Mohanty, Morris Yau
  FOCS 2020 [ArXiv]

• Optimal Mean Estimation without a Variance
  Yeshwanth Cherapanamjeri, Nilesh Tripuraneni, Peter L. Bartlett, Michael I. Jordan
  Extended Abstract at COLT 2022 [ArXiv]

• Algorithms for Heavy-Tailed Statistics: Regression, Covariance Estimation, and Beyond
  Yeshwanth Cherapanamjeri, Samuel B. Hopkins, Tarun Kathuria, Prasad Raghavendra, Nilesh Tripuraneni
  STOC 2020 [ArXiv]

• Fast Mean Estimation with Sub-Gaussian Rates
  Yeshwanth Cherapanamjeri, Nicolas Flammarion, Peter L. Bartlett
  COLT 2019 [ArXiv]

• Testing Markov Chains without Hitting
  Yeshwanth Cherapanamjeri, Peter L. Bartlett
  COLT 2019 [ArXiv]

• Thresholding based Efficient Outlier Robust PCA
  Yeshwanth Cherapanamjeri, Prateek Jain, Praneeth Netrapalli
  COLT 2017 [ArXiv]

• Nearly Optimal Robust Matrix Completion
  Yeshwanth Cherapanamjeri, Kartik Gupta, Prateek Jain
  ICML 2017 [ArXiv]