Brian Kulis

University of California, Berkeley

Papers

1

Total Citations

19

H-Index

1

About

Brian Kulis is a leading researcher in machine learning, with a primary focus on metric learning, reinforcement learning, and representation learning. His work bridges the gap between learning effective data representations and enabling intelligent agents to make decisions in complex environments. Kulis is best known for pioneering metric learning techniques that allow algorithms to automatically discover similarity and distance functions from data, a foundational contribution that has influenced fields from computer vision to robotics. His influential 2011 paper, "Metric Learning for Reinforcement Learning Agents," which has garnered 19 citations, introduced a novel framework for learning state representations directly within reinforcement learning systems—moving beyond hand-coded features to adaptive, learned representations. This work demonstrated how metric learning could enhance an agent’s ability to generalize across tasks, a critical step toward more autonomous and efficient AI. Kulis’s research has had a lasting impact on how modern machine learning systems perceive and interact with their environments, making him a key figure in the evolution of adaptive, representation-driven algorithms.

Research Focus

Key Achievements

1
H-Index
1
Papers
19
Total Citations
19
Avg Citations/Paper
🏆 Most Cited Paper
Metric learning for reinforcement learning agents
19 citations · 2011
📈 Most Prolific Year: 2011 (1 Papers)
🤝 Key Collaborators: 2
🏛 Institutions: University of California, Berkeley

Top Papers

  1. 1

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 11 days ago