AIConfPaper
← Search

Dheeraj Mysore Nagaraj

14 accepted papers

2026

Fine-Tuning Diffusion Models via Intermediate Distribution Shaping

Gautham Govind Anil, Shaan Ul Haque, Nithish Kannen, Dheeraj Mysore Nagaraj, Karthikeyan Shanmugam, Sanjay Shakkottai
ICLR 2026poster

Diffusion models are widely used for generative tasks across domains. While pre-trained diffusion models effectively capture the training data distribution, it is often desirable to shape these distributions using reward functions to align with downstream applications. Policy gradient methods, such…

Cited by 0Source ↗Scholar ↗
2026

Poisson Midpoint Method for Log Concave Sampling: Beyond the Strong Error Lower Bounds

Rishikesh Srinivasan, Dheeraj Mysore Nagaraj
ICLR 2026poster

We study the problem of sampling from strongly log-concave distributions over $\mathbb{R}^d$ using the Poisson midpoint discretization (a variant of the randomized midpoint method) for overdamped/underdamped Langevin dynamics. We prove its convergence in the 2-Wasserstein distance ($\mathcal W_2$),…

Cited by 0Source ↗Scholar ↗
2025

Dimension-free Score Matching and Time Bootstrapping for Diffusion Models

Syamantak Kumar, Dheeraj Mysore Nagaraj, Purnamrita Sarkar
NeurIPS 2025poster

Diffusion models generate samples by estimating the score function of the target distribution at various noise levels. The model is trained using samples drawn from the target distribution, progressively adding noise. Previous sample complexity bounds have a polynomial dependence on the dimension $d…

Cited by 0Source ↗Scholar ↗
2025

Glauber Generative Model: Discrete Diffusion Models via Binary Classification

Harshit Varma, Dheeraj Mysore Nagaraj, Karthikeyan Shanmugam
ICLR 2025poster

We introduce the Glauber Generative Model (GGM), a new class of discrete diffusion models, to obtain new samples from a distribution given samples from a discrete space. GGM deploys a discrete Markov chain called the heat bath dynamics (or the Glauber dynamics) to denoise a sequence of noisy tokens…

Cited by 3Source ↗PDF ↗Scholar ↗
2025

The Bandit Whisperer: Communication Learning for Restless Bandits

Yunfan Zhao, Tonghan Wang, Dheeraj Mysore Nagaraj, Aparna Taneja, Milind Tambe
AAAI 2025technical

Applying Reinforcement Learning (RL) to Restless Multi-Arm Bandits (RMABs) offers a promising avenue for addressing allocation problems with resource constraints and temporal dynamics. However, classic RMAB models largely overlook the challenges of (systematic) data errors - a common occurrence in r…

Cited by 6Source ↗PDF ↗Scholar ↗
2024

A Decision-Language Model (DLM) for Dynamic Restless Multi-Armed Bandit Tasks in Public Health

Nikhil Behari, Edwin Zhang, YUNFAN ZHAO, Aparna Taneja, Dheeraj Mysore Nagaraj, Milind Tambe
NeurIPS 2024poster

Restless multi-armed bandits (RMAB) have demonstrated success in optimizing resource allocation for large beneficiary populations in public health settings. Unfortunately, RMAB models lack flexibility to adapt to evolving public health policy priorities. Concurrently, Large Language Models (LLMs) ha…

Cited by 14Source ↗PDF ↗Scholar ↗
2024

Near-Optimal Streaming Heavy-Tailed Statistical Estimation with Clipped SGD

Aniket Das, Dheeraj Mysore Nagaraj, Soumyabrata Pal, Arun Suggala, Prateek Varshney
NeurIPS 2024poster

$\newcommand{\Tr}{\mathsf{Tr}}$ We consider the problem of high-dimensional heavy-tailed statistical estimation in the streaming setting, which is much harder than the traditional batch setting due to memory constraints. We cast this problem as stochastic convex optimization with heavy tailed stocha…

Cited by 2Source ↗PDF ↗Scholar ↗
2024

The Poisson Midpoint Method for Langevin Dynamics: Provably Efficient Discretization for Diffusion Models

Saravanan Kandasamy, Dheeraj Mysore Nagaraj
NeurIPS 2024poster

Langevin Dynamics is a Stochastic Differential Equation (SDE) central to sampling and generative modeling and is implemented via time discretization. Langevin Monte Carlo (LMC), based on the Euler-Maruyama discretization, is the simplest and most studied algorithm. LMC can suffer from slow convergen…

Cited by 6Source ↗PDF ↗Scholar ↗
2023

Multi-User Reinforcement Learning with Low Rank Rewards

Dheeraj Mysore Nagaraj, Suhas S Kowshik, Naman Agarwal, Praneeth Netrapalli, Prateek Jain
ICML 2023poster

We consider collaborative multi-user reinforcement learning, where multiple users have the same state-action space and transition probabilities but different rewards. Under the assumption that the reward matrix of the $N$ users has a low-rank structure -- a standard and practically successful assump…

Cited by 1Source ↗PDF ↗Scholar ↗
2023

Provably Fast Finite Particle Variants of SVGD via Virtual Particle Stochastic Approximation

Aniket Das, Dheeraj Mysore Nagaraj
NeurIPS 2023spotlight

Stein Variational Gradient Descent (SVGD) is a popular particle-based variational inference algorithm with impressive empirical performance across various domains. Although the population (i.e, infinite-particle) limit dynamics of SVGD is well characterized, its behavior in the finite-particle regim…

Cited by 13Source ↗PDF ↗Scholar ↗
2022

Online Target Q-learning with Reverse Experience Replay: Efficiently finding the Optimal Policy for Linear MDPs

Naman Agarwal, Syomantak Chaudhuri, Prateek Jain, Dheeraj Mysore Nagaraj, Praneeth Netrapalli
ICLR 2022poster

Q-learning is a popular Reinforcement Learning (RL) algorithm which is widely used in practice with function approximation (Mnih et al., 2015). In contrast, existing theoretical results are pessimistic about Q-learning. For example, (Baird, 1995) shows that Q-learning does not converge even with lin…

Cited by 31Source ↗PDF ↗Scholar ↗
2021

Near-optimal Offline and Streaming Algorithms for Learning Non-Linear Dynamical Systems

Suhas S Kowshik, Dheeraj Mysore Nagaraj, Prateek Jain, Praneeth Netrapalli
NeurIPS 2021spotlight

We consider the setting of vector valued non-linear dynamical systems $X_{t+1} = \phi(A^{*} X_t) + \eta_t$, where $\eta_t$ is unbiased noise and $\phi : \mathbb{R} \to \mathbb{R}$ is a known link function that satisfies certain {\em expansivity property}. The goal is to learn $A^{*}$ from a single t…

Cited by 39Source ↗PDF ↗Scholar ↗
2021

Streaming Linear System Identification with Reverse Experience Replay

Suhas S Kowshik, Dheeraj Mysore Nagaraj, Prateek Jain, Praneeth Netrapalli
NeurIPS 2021poster

We consider the problem of estimating a linear time-invariant (LTI) dynamical system from a single trajectory via streaming algorithms, which is encountered in several applications including reinforcement learning (RL) and time-series analysis. While the LTI system estimation problem is well-studie…

Cited by 21Source ↗PDF ↗Scholar ↗
2021

The staircase property: How hierarchical structure can guide deep learning

Emmanuel Abbe, Enric Boix-Adserà, Matthew Stewart Brennan, Guy Bresler, Dheeraj Mysore Nagaraj
NeurIPS 2021poster

This paper identifies a structural property of data distributions that enables deep neural networks to learn hierarchically. We define the ``staircase'' property for functions over the Boolean hypercube, which posits that high-order Fourier coefficients are reachable from lower-order Fourier coeffic…

Cited by 68Source ↗PDF ↗Scholar ↗