Adarsh Prasad

I am a graduate student(PhD) in the Machine Learning Department at Carnegie Mellon University, where I am advised by Pradeep Ravikumar and Sivaraman Balakrishnan. Previously, I got my Bachelors degree in Electrical Engineering at IIT Delhi. While at IIT, I worked with Manik Varma on Sparse Multilabel Learning. I am broadly interested in statistical machine learning, high-dimensional statistics and optimization. Most recently, I have been working on computationally efficient methods for robust statistics, and their application to safe and reliable machine learning. My PhD is generously supported by JP Morgan AI Research PhD Fellowship. Apart from theoretical research, I enjoy working with real-world data, particularly noisy data. I represented CMU at Citadel Datathon Finals, and have spent some time interning in quantitative trading at Cubist Systematic Strategies. I am also an avid squash player. Email :

   Publications

   Preprints

• A Unified Approach to Robust Mean Estimation [pdf]
     Adarsh Prasad, Sivaraman Balakrishnan, Pradeep Ravikumar.
     Under Submission

   Journal Papers

• Robust Estimation via Robust Gradient Estimation [pdf]
     Adarsh Prasad, Arun Sai Suggala, Sivaraman Balakrishnan, Pradeep Ravikumar.
     Journal of the Royal Statistical Society: Series B (Statistical Methodology) (JRSSB), 2020.

   Conference Papers

• Robust Linear Regression: Optimal Rates in Polynomial Time [pdf]
     Ainesh Bakshi* and Adarsh Prasad*
     Symposium on Theory of Computing (STOC) 2021 (To Appear)

• On Learning Ising Models under Huber's Contamination Model. [main ] [supp ]
     A.Prasad*, V. Srinivasan*, S. Balakrishnan, P. Ravikumar
     Neural Information Processing Systems (NeurIPS) 2020.

• A Robust Univariate Mean Estimator is All You Need. [main ] [supp]
     Adarsh Prasad, Sivaraman Balakrishnan, Pradeep Ravikumar.
     In International Conference on Artificial Intelligence and Statistics (AISTATS) 23, 2020

• Uniform Convergence of Rank-Weighted Learning. [pdf]
  J. Khim, L. Liu, A. Prasad, P. Ravikumar
     In International Conference on Machine Learning (ICML 2020)

• Revisiting Adversarial Risk [pdf]
     A. Suggala, A. Prasad, V. Nagarajan, P. Ravikumar.
     In International Conference on Artificial Intelligence and Statistics (AISTATS), 2019

• Connecting Optimization and Regularization Paths [pdf]
     A. Suggala, A. Prasad, P. Ravikumar.
     In Advances in Neural Information Processing Systems (NIPS), 2018

• On Separability of Loss Functions, and Revisiting Discriminative Vs Generative Models [pdf]
     A. Prasad, A. Niculescu-Mizil, P. Ravikumar.
     In Advances in Neural Information Processing Systems (NIPS), 2017  (Spotlight)

• Fast Classification Rates for High-dimensional Conditional Gaussian Models [pdf]
     T. Li, A. Prasad, P. Ravikumar
     In Advances in Neural Information Processing Systems (NIPS), 2015

• Distributional Rank Aggregation, and an Axiomatic Analysis [pdf] [Supp]
     A. Prasad*, H. Pareek*, P. Ravikumar
     In International Conference on Machine Learning (ICML) 32, 2015

• Submodular meets Structured: Finding Diverse Subsets in Exponentially-Large Structured Item Sets [pdf] [Data]
     A. Prasad, S. Jegelka, D. Batra
     In Advances in Neural Information Processing Systems (NIPS), 2014  (Spotlight)

*: Equal Contribution or Alphabetical Order