Jordi Grau-Moya

Max Planck Society

Papers

2

Total Citations

11

H-Index

2

About

Jordi Grau-Moya is a leading researcher at the intersection of reinforcement learning (RL) and information theory, whose work redefines how agents explore and learn. His primary research areas include intrinsic motivation, entropy regularization, and the development of principled frameworks for decision-making under uncertainty. Grau-Moya’s major contribution is the introduction of a unified Bellman optimality principle that seamlessly combines reward maximization with empowerment—an information-theoretic measure of an agent’s control over its environment. This groundbreaking work, cited 8 times, demonstrates that by encouraging agents to visit states with a high number of reachable next states, empowerment can drive complex, self-directed exploration without external rewards. Additionally, his research on mutual-information regularization in Markov decision processes (cited 3 times) formalizes how cumulative entropy regularization promotes high-entropy policies, effectively improving exploration and robustness in actor-critic learning. Grau-Moya’s insights are pivotal for advancing autonomous systems, offering a rigorous mathematical foundation for building agents that are not only reward-driven but also intrinsically curious and adaptable. His work continues to inspire new directions in safe and exploratory AI.

Research Focus

Key Achievements

2
H-Index
2
Papers
11
Total Citations
6
Avg Citations/Paper
🏆 Most Cited Paper
A Unified Bellman Optimality Principle Combining Reward Maximization and Empowerment
8 citations · 2019
📈 Most Prolific Year: 2019 (2 Papers)
🤝 Key Collaborators: 2
🏛 Institutions: Max Planck Society

Top Papers

  1. 1
  2. 2

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 15 days ago