Aswin Paul

About

Aswin Paul is a physicist turned machine learning researcher with a PhD from the joint IIT Bombay–Monash University program, where his thesis focused on developing computationally efficient and explainable approaches to active inference. His research spans brain-inspired algorithms, efficiency, and interpretability in machine learning, with publications in venues such as Expert Systems With Applications and Entropy. He currently works at VERSES AI in collaboration with Prof. Christopher Buckley, building on prior collaborations with Karl Friston’s group. His background combines rigorous physics training with expertise in probabilistic modelling and decision-making under uncertainty, supported by multiple academic awards and peer-reviewed contributions.

Employment

• VERSES (United States) Researcher
  2025 - Present

Education

• Monash University PhD
  2019 - 2024

Projects & Funding

Projects & funding information is unavailable.

Publications (5)

• Dynamic Network Plasticity and Sample Efficiency in Biological Neural Cultures: A Comparative Study with Deep Reinforcement Learning
  Cyborg and Bionic Systems 2025 DOI: 10.34133/cbsystems.0336
  +Save
• On efficient computation in active inference
  Expert Systems with Applications 2024 DOI: 10.1016/j.eswa.2024.124315
  +Save
• On Predictive Planning and Counterfactual Learning in Active Inference
  Entropy 2024 DOI: 10.3390/e26060484
  +Save
• Active Inference for Stochastic Control
  2021
  +Save
• Active Inference for Stochastic Control
  Communications in Computer and Information Science 2021 DOI: 10.1007/978-3-030-93736-2_47
  +Save