...

Hi, I’m Arjun Vaithilingam Sudhakar 👋

I am in the final year of my Research master’s at Mila-Quebec AI Institute under the supervision of Professor Sarath Chandar in Montreal, Canada. I also hold a Master’s degree from Mila and the University of Montreal, where I focused on language models. Before my Research Master, I worked as a Machine Learning Engineer for three years at Wipro and as a AI Research Intern at Hydro Quebec, a government organization in Quebec, Canada.

Research Focus: My research centers on the intersection of Reinforcement Learning (RL) and Large Language Models (LLM). While RL exhibits sample inefficiency, LLM boasts general-purpose knowledge but needs to improve in planning and sequential decision-making. My objective is to seamlessly integrate both paradigms to enhance the capabilities of general-purpose agents, equipping them with superior decision-making prowess.

Research Interest : Reinforcement learning, Natural language processing

Links : Google Scholar , Linkedin , Github

Education

...
Korbit, Montreal, Quebec
Job Title: Machine Learning Specialist

Timeline: March 2024 - Present

...
Huawei, Montreal, Quebec
Job Title: AI Research Assistant

Timeline: March 2023 - September 2023

...
Mila - Quebec AI Institute, Chandar Lab
Job Title: AI Research Assistant

Timeline: August 2021 - August 2022

...
Hydro Quebec, Montreal, Quebec
Job Title: AI Research Intern

Timeline: May 2021 - January 2022

...
Wipro Technologies, India
Job Title: Machine Learning Engineer

Timeline: October 2017 - July 2020

...
Language Model-In-The-Loop Data Optimal Approach to Learn-To-Recommend Actions in Text Games
Arjun Vaithilingam Sudhakar, Prasanna Parthasarathi, Janarthanan Rajendran, Sarath Chandar

ICML 2024 Workshop on Foundation Models in the Wild, 2024

...
Multi-agent text-based Hanabi challenge
H Nekoei*, Arjun Vaithilingam Sudhakar*, J Rajendran, M Liu, S Chandar

ICLR 2024, Generative Models for Decision Making Workshop

...
Feature diversity in self-supervised learning
Pranshu Malviya*, Arjun Vaithilingam Sudhakar*

Conference on Lifelong Learning Agents (CoLLAs)-2022 - Workshop Track

...
Multi Label Deep Learning classification approach for False Data Injection Attacks in Smart Grid.
Prasanna Srinivasan V., Balasubadra K., Saravanan K, Arjun Vaithilingam Sudhakar, Malarkodi S.

KSII Transactions on Internet & Information Systems - Journal

Lab Manager, Chandar Research Lab, Mila - Quebec AI Institute

August 2021 - April 2024

  • Oversaw successful submission of proposals like Google teaching proposal and CIFAR.
  • Improved lab functioning through new systems.
  • Formed a cost-effective web development team through outsourcing, saving research time and increasing lab visibility.
  • Local Area Chair, CoLLAs2023

    December 2022 - August 2023

  • Overseen and synchronized all aspects of the conference, including managing meetings, planning, and executing the event.
  • Additionally, I lead a team of 5 to ensure a smooth and successful outcome.
  • Organizing Committee, CoLLAs2022

    December 2021 - August 2022

  • I efficiently managed meetings, accommodation, website, and dining arrangements to ensure a seamless and well-coordinated conference experience.
  • Toastmasters International (Nonprofit Educational Organization)

    April 2018 - Present

  • Led 150+ members in District 92 of Toastmasters Int'l, managing multiple business units across different locations.
  • Persuaded CXO to implement a $25,000 USD/yr membership reimbursement program (pilot) across Wipro.
  • Increased club strength from 4 to 40 members in under a year, and organized a 100+ audience milestone event and club contests for public speaking.
  • Teaching Assistant at AI4Good - Machine Learning

    June 2022 - July 2022

  • Guided and mentored underprivileged female students in Machine/Deep Learning, teaching math, ML/DL/RL.
  • Mentored capstone project on Emotive Application using ML/DL over 4 weeks.
  • Teaching Assistant at Polytechnique Montreal - Machine Learning- INF8245E

    September 2021 - December 2021

  • Proposed and led math and python tutorials for students, including linear algebra and scikit.
  • Delivered lectures in-class and online. Kept track of student feedback for course improvement.