Pranav Agarwal

Pranav Agarwal

About Me

I am a graduate reserach student at MILA and ETS Montreal, working on deep reinforcement learning and robotics, supervised by Prof. Samira Ebrahimi Kahou and Prof. Sheldon Andrews . I am also a Research Intern at CM-Labs. Previously, I was working on Autonomous Driving at INRIA (RITS), supervised by Dr. Raoul de CHARETTE . In the past, I worked as a collaborator with Prof. ‪Natalia Díaz-Rodríguez (INRIA, Flowers). My research interest center around Deep Reinforcement Learning for Robotic manipulation and Deep Learning for Vision based applications.

I completed my Bachelor’s in Electronics and Communication Engineering from Indian Institute of Information Technology, Guwahati, where I was awarded the President’s Gold medal for being the overall topper. During the course, I was fortunate to work as a research intern under Prof. Gemma Roig at Singapore University of Technology and Design.

Interests

  • Deep Reinforcement Learning
  • Deep Learning
  • Robotics
  • Autonomous Driving
  • Tiny Machine Learning
  • Generative Networks

Education

  • Graduate Student, 2022-2025

    MILA, Montreal

  • B.Tech in Electronics and Communication Engineering, 2015-2019

    Indian Institute of Information Technology Guwahati

Experience

 
 
 
 
 

Research Student (Masters)

MILA

Jan 2022 – Present Montreal
  • Research Areas : Reinforcement Learning, Robotics, Computer Vision, Autonomous Driving
 
 
 
 
 

Research Intern

CM-Labs

Jan 2022 – Present Montreal
  • Reinforcement Learning for Crane Manipulation in a simulated environment
 
 
 
 
 

Research Assistant

INRIA

Aug 2019 – Mar 2021 Paris
  • Worked on Reinforcement Learning (RL) Algorithms (DDPG, TD3 and PPO) for Autonomous Driving.
  • Implemented an OpenAI Gym like wrapper for CARLA Simulator to train and test different RL algorithms.
  • Proposed a novel curriculum driven multi policy RL agent to learn to drive using only sparse rewards.
 
 
 
 
 

Research Collaborator

INRIA

May 2019 – Apr 2020 Paris
  • Worked on Image Captioning (IC) Algorithms (YOLO, NOC and DNOC) to caption Egoshots dataset.
  • Proposed a new IC metric, Semantic Fidelity to evaluate diversity in image captioning models.
 
 
 
 
 

Research Intern

Singapore University of Technology and Design

May 2019 – Apr 2020 Singapore
  • Worked on Eccentricity Convolution Neural Network (ECNN).
  • Compared the performance of ECNN on ImageNet and FaceScrub to the existing models like AlexNet.

Accomplishments

  • MITACS Accelerate Fellowship for Graduate Studies.
  • Awarded the President’s Gold Medal for graduating with highest GPA.
  • Received the best Technology award by Government of India at Vibrant Gujrat.
  • Merit certificate CBSE XII exams. (full marks - top 0.1%)