About Me

I am a rising 5th year PhD student at UC Berkeley, jointly advised by Jiantao Jiao and Kannan Ramchandran, affiliated with the BLISS and BAIR labs. I was previously an intern with Nevena Lazic and Dong Yin at Deepmind and with Ravishankar Krishnaswamy at MSR India.

I work on a variety of topics in the theory of machine learning with a general focus on the statistical and computational aspects of adaptive decision making problems and reinforcement learning. More recently, I have been interested in the application of these techniques in pushing our understanding of large language models. My research has largely focused on using mathematical frameworks to explain curious practical phenomena and ultimately provide intuitions missing in existing approaches.

I am currently an organizer of the BLISS seminar and CLIMB seminar at Berkeley. Shoot me an email if you are interested in giving a talk at either venue!

In a previous life, I was a dual degree student at the Department of Electrical Engineering, IIT Madras. I am fortunate to have had Andrew Thangaraj as my thesis advisor and to have worked closely with Rahul Vaze.

Updates

April 2023: I taught a guest lecture for Statistical Machine Learning (ECE 6254) at Georgia Tech.

April 2023: I presented our work on pruning for matrix sensing at Georgia Tech.

January 2023: I presented our recent work on nonlinear bandits at Google Research India and MSR India.

November 2022: I will be at Neurips 2022. Hit me up if you will be around!

October 2022: I gave a talk on nonlinear bandits at the WNCG group at UT Austin.

October 2022: I will be participating in AIDS LifeCycle 2023. You can support me by clicking here. Every little bit counts!

Publications

Beyond UCB: Statistical complexity and optimal algorithms for non-linear ridge bandits.

Nived Rajaraman, Yanjun Han, Jiantao Jiao, Kannan Ramchandran

PDF Submitted to the Annals of Statistics

Greedy pruning with group lasso provably generalizes for matrix sensing.

Nived Rajaraman, Devvrit, Aryan Mokhtari, and Kannan Ramchandran

PDF Submitted to NeurIPS 2023

Sample efficient deep reinforcement learning via local planning.

Dong Yin, Sridhar Thiagarajan, Nevena Lazic, Nived Rajaraman, Botao Hao, and Csaba Szepesvari

PDF

Minimax optimal online imitation learning via replay estimation.

Gokul Swamy*, Nived Rajaraman*, Matt Peng, Sanjiban Choudhury, J. Bagnell, Steven Z. Wu, Jiantao Jiao, and Kannan Ramchandran (* = equal contribution)

NeurIPS 2022

PDF Code

Semi-supervised active linear regression.

Nived Rajaraman, Devvrit, and Pranjal Awasthi

NeurIPS 2022

PDF

Spectral regularization allows data-frugal learning over combinatorial spaces.

Amirali Aghazadeh, Nived Rajaraman, Tony Tu, and Kannan Ramchandran

TMLR

PDF Code

On the value of interaction and function approximation in imitation learning.

Nived Rajaraman, Yanjun Han, Lin Yang, Jingbo Liu, Jiantao Jiao, and Kannan Ramchandran

NeurIPS 2021

PDF

Provably breaking the quadratic error compounding barrier in imitation learning, optimally.

Nived Rajaraman, Yanjun Han, Lin F. Yang, Kannan Ramchandran, and Jiantao Jiao

PDF Submitted to the Annals of Statistics

Towards the fundamental limits of imitation learning.

Nived Rajaraman, Lin F. Yang, Jiantao Jiao, Kannan Ramchandran

NeurIPS 2020

PDF

FastSecAgg: Scalable secure aggregation for privacy-preserving federated learning.

Swanand Kadhe, Nived Rajaraman, O. Ozan Koyluoglu, and Kannan Ramchandran

ICML Workshop on FL for User Privacy and Data Confidentiality (2020); CCS Workshop on Privacy-Preserving Machine Learning in Practice (2020); ISIT (2021)

PDF

Robust correlation clustering.

Ravishankar Krishnaswamy, Devvrit, and Nived Rajaraman (equal contribution)

APPROX/RANDOM 2019

PDF