Aarshvi Gajjar

I am a final-year PhD student in Computer Science at NYU, where I am affiliated with the Theoretical Computer Science Group; and advised by Christopher Musco and Chinmay Hegde.

Before this, I completed a master's at UMass Amherst, where I wrote my thesis with Cameron Musco, and a I completed a bachelor's at IIIT Hyderabad.

I’ve spent time at Goldman Sachs and Amazon, and most recently interned at Netflix and Uber.

Research

I study principled methods for choosing queries to maximize a machine learning model's performance with limited data. This spans active learning, optimal experiment design, efficient exploration and decision making.

I use tools from theoretical computer science, statistical learning theory and high dimensional statistics to prove guarantees.

Directions I am intersted in:

Selecting a small subset of training-data to fine-tune models using few labeled samples.
Active regression under safety constraints.
Links between pure exploration and techniques from randomized numerical linear algebra.

Publications

Constrained Active Regression via Ellipsoid Regularized Leverage Scores
Aarshvi Gajjar, Syamantak Kumar, Christopher Musco, Yihan Zhou
Preliminary Version: Constrained ML Workshop, NeurIPS 2025

Putting the Spotlight on Initial State Distribution
Aditya Makkar, Aarshvi Gajjar and Eugene Vinitsky
Preliminary Version: Finding the Frame Workshop, RLC, 2025

Agnostic Active Learning of Single Index Models with Linear Sample Complexity
†Aarshvi Gajjar, †Wai Ming Tai, †Xingyu Xu, Chinmay Hegde, Christopher Musco and Yi Li
COLT, 2024
Associated poster for minisymposium on Scientific ML for Scarce Data, SIAM MDS24
Preliminary version: Adaptive Experimental Design and Active Learning Workshop, NeurIPS, 2023

Active Learning for Single Neuron Models with Lipschitz Non-Linearities
Aarshvi Gajjar, Chinmay Hegde and Christopher Musco
AISTATS, 2023
Preliminary version: selected as Spotlight at DLDE Workshop , NeurIPS 2022

Subspace Embeddings under Nonlinear Transformations
Aarshvi Gajjar, Cameron Musco
ALT, 2021

† indicates equal contribution

Contact: [first-name]@nyu.edu.

Other

Outside research, I have varied interests -- fitness, hikes both urban and in nature. I also try to make time to read; lately my go-to authors have been Bertrand Russell and Gene Wolfe.