Learning Agents Research Collective (LARC)
Welcome to LARC!
At the intersection of Artificial Intelligence, Control Theory, and Collective Intelligence, our research group focuses on engineering AI systems that can learn, adapt, and cooperate in complex, dynamic environments. Our goal is to advance the theoretical foundations of reinforcement learning while developing scalable, practical algorithms for real-world impact. From optimizing critical infrastructure to seamless Human-AI collaboration, we design the intelligent systems of tomorrow.
Core Research Themes:
- Reinforcement Learning (RL): We investigate the foundational theory and algorithmic design of Single-Agent RL.
- Key Areas: Sample efficiency, Safe RL, Constrained RL, Teacher-Student Paradigm, Sparse-Reward, Hierarchical RL, etc.
- Multi-Agent Learning: We develop algorithms that allow multiple agents to learn simultaneous strategies, balancing competition and cooperation without global oversight.
- Key Areas: Large-Scale Multi-Agent Systems, Adhoc RL, Multi-Task Multi-Agent RL, Multi-Drone Systems, etc.
- Human-AI Interaction: We study how AI systems can learn from, adapt to, and safely collaborate with human users.
- Key Areas: Human-in-the-loop Reinforcement Learning, Social Reinforcement Learning, etc
Current Real-World Application Areas:
- Smart Energy Management: Designing multi-agent frameworks for smart grid optimization, and efficient resource distribution.
- Smart Transportation: Developing robust collective decision-making algorithms for smart traffic managment.
- Drone Swarms: Developing robust collective decision-making algorithms for coordinated drone swarms.
Computing Resources:
- 96 Cores AMD Threadripper, 312 GB Memory, NVIDIA RTX 5000 32 GB Ada Gen.
- 96 Cores AMD Threadripper, 215 GB Memory, NVIDIA RTX PRO 6000 Blackwell GPU, 96 GB.
- Access to IIT Delhi PARAM PRAGYA - National Supercomputing Facility - 400, A100. GPUs
- Access to IIT Delhi PADUM - High Performance Computing Facility
Team Members:
- James Arambam, Assistant Professor, Yardi School of AI, IIT Delhi
Graduate Researchers:
- Abhay Singh, MTech MINDS, ScAI - Master Thesis Project (MTP)
- Yuvraj Verma, MTech MINDS, ScAI - MTP
- Kashish Srivastava, MTech MINDS, ScAI - MTP
- Joel M, MTech CSE, MTP
- Soumya Namdeo, Int. MTech, Math & Computing - MTP
Undergraduate Researchers:
- Shubham Shukla, BTech Mech. - Bachelor Thesis Project (BTP)
- Saumitra Garg, BTech CSE - BTP
- Anubhav Kar, BTech CSE - BTP
- Abhishek Amrendra Kumar, BTech CSE - BTP
- Rohan Roy, BTech Math & Computing - BTP
- Sarthak Maheswari, BTech Math & Computing - BTP
- Sikhar Gupta, BTech Math & Computing - BTP
- Abhimanyu Goyal, BTech Mech. Engg.
- Shubhankur Tripathi, BTech Elec. Engg.
Past Members:
- Rishit Jhakaria, BTech CSE - BTP
- Raj Aryan, BTech Math & Computing - BTP
- Shamil Mohammed, BTech Math & Computing - BTP
- Dhruva Bhardwaj, MTech Robotics, Elec. Engg. - Minor Project.
Join Us:
We are always looking for passionate students, postdocs, and industry collaborators interested in pushing the boundaries of Reinforcement Learning and Multi-Agent Learning. If interested, reach out directly via email: rlgroup.iitd@gmail.com with your CV and a brief statement of research interests.