M Ganesh Kumar
Also known as: M Ganeshkumar
Max Planck Institute for Biological Cybernetics, Harvard University, National University of Singapore, Institute of High Performance Computing
0000-0001-5559-6428
About
I am a Postdoctoral Fellow at the Max Planck Institute of Biological Cybernetics. I completed my first postdoctoral training at the School of Engineering and Applied Sciences at Harvard University. I earned my PhD in Computational Neuroscience at the National University of Singapore. I graduated from the National University of Singapore with a Life Sciences degree with Honors.
How do biological neural circuits and learning rules decide the next best action? What mathematical model can help us understand these phenomena? How we improve existing artificial systems and devise tools to alleviate disorders?
To explore these questions, I develop artificially intelligent agents, grounded to theory and experiments, to understand learning computations for intelligence, and when it might fail. Based on these insights, I hope to improve existing artificial systems, and develop technologies that can improve learning outcomes and alleviate learning disabilities.
Employment
-
Max Planck Institute for Biological Cybernetics Postdoctoral Fellow2026 - Present
-
Harvard University Postdoctoral Fellow2023 - 2025
-
Institute of High Performance Computing Scientist 12022 - 2023
-
Agency for Science, Technology and Research (A*STAR) Research Engineer2017 - 2018
Education
-
National University of Singapore Ph.D2018 - 2022
-
National University of Singapore Bachelor of Science (Honors - Distinction)2013 - 2017
Projects & Funding
Projects & funding information is unavailable.
Publications (15)
- CompleteP for RL: Maintaining Feature Learning When Scaling Deep Reinforcement Learning Save
- Author Correction: Predictive coding of reward in the hippocampus Save
- Predictive coding of reward in the hippocampus Save
- Neurocomputational Underpinnings of Suboptimal Beliefs in Reinforcement Learning Agents Save
- A Model of Place Field Reorganization During Reward Maximization Save
- HUMAN-LIKE COMPOSITIONAL LEARNING OF VISUALLY-GROUNDED CONCEPTS USING SYNTHETIC ENVIRONMENTS Save
- Neurocomputational Underpinnings of Suboptimal Beliefs in Reinforcement Learning Agents Save
- Neurocomputational underpinnings of suboptimal beliefs in recurrent neural network-based agents Save
- Trends Innovations Challenges in Employing Interdisciplinary Approaches to Biomedical Sciences Save
- DetermiNet: A Large-Scale Diagnostic Dataset for Complex Visually-Grounded Referencing using Determiners Save