Email:  ayushc@mit.edu

  •     •     •  

About Me

I am a PhD student at MIT Media Lab supervised by Prof. Ramesh Raskar. My research makes theoretical advances in multi-agent simulations and distributed computation. My work has been published in leading venues like AAMAS, CVPR, and BMJ, received multiple best paper awards, resulted in 25 patents and has been deployed across several countries. Prior to MIT, I was an engineer at Adobe where I received the Outstanding Young Engineer Award for my work on collaborative systems. To learn more about my research:

Research Overview

I develop Large Population Models , a new computational framework that enables simulating and understanding how millions of individual agents combine to shape societal outcomes. LPMs overcome limits of traditional agentic simulation via three key innovations:

  1. Scalable: Simulate 8M+ agents with sophisticated behavior one a single GPU(600× faster)
  2. Differentiable: Calibrate to real-world data by optimizing through large compute graphs with mechanistic logic and multi-scale dynamics (8300× speedup)
  3. Decentralized: Integrate real-time individual data and backpropagate through reality via novel cryptographic protocols
AgentTorch, our open-source framework, makes building and deploying massive LPMs accesible. LPMs are making real-world impact: optimizing vaccine distribution strategies to help immunize millions, and improving sustainability of billion dollar supply chains globally.
Publications   ( show highlighted / show all by date / show by topic )
Keywords:  Agent-based Simulations • Decentralized Computation •  Machine Learning •  Applications 
(* indicates equal contribution)
On the limits of agency in agent-based models
Ayush Chopra , Shashank Kumar , Nurullah Giray Kuru , Ramesh Raskar , Arnau Quera-bofarull
Accepted at 24th Autonomous Agents and Multi-agent Systems AAMAS , 2025
[PDF]
Differentiable Agent-based Modeling: Systems, Method and Applications
Ayush Chopra , Arnau Quera-bofarull , Sijin Zhang
Accepted at 23rd Autonomous Agents and Multi-agent Systems AAMAS , 2024
[Slides] • [Tutorial]
Private Agent-based Modeling
Ayush Chopra , Arnau Quera-bofarull , Nurullah Giray Kuru , Michael Wooldridge , Ramesh Raskar
Accepted at 23rd Autonomous Agents and Multi-agent Systems AAMAS , 2024
[Preprint]
flame: a Framework for Learning in Agent-based Models
Ayush Chopra , Jayakumar Subramanian , Ramesh Raskar
Accepted at 23rd Autonomous Agents and Multi-agent Systems AAMAS , 2024
[PDF]
Differentiable Agent-based Epidemiology
Ayush Chopra , Alex Rodriguez , Jayakumar Subramanian , Aditya Prakash , Ramesh Raskar
Accepted at 22nd Autonomous Agents and Multi-agent Systems AAMAS , 2023
[PDF] Best Paper Award ICML Workshop 2022
Distributed Calibration of Agent-based Models
Aditi Garg , Ayush Chopra
7th International workshop on Epidemiology meets Data Mining and Knowledge discovery (KDD) epiDAMIK , 2023
[Preprint]
Powering Robust Fashion Retrieval With Information Rich Feature Embeddings
Ayush Chopra , Abhishek Sinha , Hiresh Gupta , Mausoom Sarkar , Balaji K
Computer Vision and Pattern Recognition Workshop CVPR Workshop , 2019
[PDF] Best Paper Award CVPR Workshop 2019

Talks
  • [2025/02] Invited Talk @ Oak Ridge National Laboratory
  • [2025/01] Invited Talk @ University of Delhi
  • [2024/10] Invited Talk @ MIT PKG Center
  • [2024/10] Invited Talk @ Rising Stars: MIT Media Lab
  • [2024/09] Invited Talk @ Google DeepMind
  • [2024/09] Invited Talk @ University of Michigan Ann Arbor
  • [2024/09] Invited Talk @ Amazon Web Services
  • [2024/09] Invited Talk @ MIT Decentralized AI Summit
  • [2024/08] Invited Talk @ Google Cambridge
  • [2024/09] Tutorial @ AAMAS 2024
  • [2023/11] Invited Talk @ Government of New Zealand
  • [2023/08] Invited Talk @ University of Vermont
  • [2023/08] Invited Talk @ ACM Goa
  • [2023/07] Invited Talk @ Columbia University Public Health
  • [2023/04] Invited Talk @ JP Morgan AI Research
  • [2022/10] Invited Talk @ University of Oxford
Academic Talk (45 minutes)
Enterprise Pitch (5 minutes)

Honors and Awards

Professional Service

Design and source code modifed from Yuzhe Yang's webpage with permission.