Arnav Choudhry

About

I am starting my fourth year (Jan 2022) as a PhD student at the Mobility Data Analytics Center (affiliated with the Civil Engineering and Machine Learning Departments), advised by Dr. Sean Qian. I am interested in using data to improve the built environment. Currently, I am working on using data to inform decision making for the next generation of last-mile delivery infrastructure. Some of my experience is related to understanding the risk and safety of machine learning algorithms used in advanced autonomous systems.

I finished my undergrad (B.Tech.) in Civil Engineering (minor in Operations Research) from IIT Madras. In the summer of 2021, I worked at Apple SPG on Data Science for autonomous systems. Earlier, I have worked as a research intern at the Research Center for Advanced Science and Technology of the University of Tokyo under Dr. Hideki Koizumi, and also as a management consulting intern at PwC India.

Research Interests

• Machine Learning: meta-learning, transfer learning, time series prediction, anomaly detection, few-shot learning
• Transportation: first/last mile delivery, smart curbspaces, computer vision applications

Publications and preprints

• Energy Consumption and Greenhouse Gas Emissions of Autonomous Robots as an Alternative for Last-mile Delivery
  Thiago A. Rodrigues, Arnav Choudhry, Natalia L. Oliveira, H. Scott Matthews, Sebastian Scherer, Constantine Samaras
  Transportation Research Board 101st Annual Meeting. TRB 2022
  [PDF]

• CVaR-based Flight Energy Risk Assessment for Multirotor UAVs using a Deep Energy Model
  Arnav Choudhry*, Brady Moon*, Jay Patrikar*, Sebastian Scherer
  The 2021 International Conference on Robotics and Automation. ICRA 2021
  [PDF] [Code]

• In-flight positional and energy use dataset of package delivery quadcopter UAVs
  Thiago Rodrigues, Jay Patrikar, Arnav Choudhry, Jacob Feldgoise, Vaibhav Arcot, Aradhana Gahlaut, Sophia Lau, Brady Moon, Bastian Wagner, H. Scott Matthews, Sebastian Scherer, and Constantine Samaras
  nature Scientific Data, 2020.
  [PDF]

• Using GPS trace data to infer goods demand at a census block level
  Arnav Choudhry, Sean Qian
  In preparation.
  [PDF] [Code] [Project]

• PhysicsVAE: Probabilistic Latent Physics Engineby Learning to Assemble Basic Knowledge
  Weiran Yao, Arnav Choudhry, Shuo Li, Chaochao Lin
  [PDF]

• Hierarchical Deep Imitative Modeling of HumanDriving Behaviors in Urban Environment
  Weiran Yao, Arnav Choudhry
  [PDF]

• Identifying irregular conditions in traffic networks using multi-modal data streams
  Arnav Choudhry
  [PDF]

Services

• Reviewer: Transportation Research – Part C and Transportation Science.
• Teaching Assistant: 12-794 Probability and Statistics, 12-401 Engineering Project Management

Powered by Jekyll and Minimal Light theme.