Diana Borsa

Google (United States)

Papers

1

Total Citations

38

H-Index

1

About

Diana Borsa is a research scientist whose work lies at the intersection of reinforcement learning, skill discovery, and hierarchical decision-making. Her most influential contribution is the introduction of the **Option Keyboard**, a framework that enables agents to combine known skills in the space of pseudo-rewards, or cumulants. This approach allows for the flexible composition of skills to solve complex, long-horizon tasks—a key challenge in modern AI. Her 2019 paper on the topic has garnered 38 citations and is widely recognized for formalizing how skills can be manipulated and recombined without retraining. Borsa’s research addresses fundamental questions in transfer learning and temporal abstraction, making her work essential for students and researchers interested in building more adaptable and efficient reinforcement learning agents. Her contributions continue to shape how the field thinks about modular skill acquisition and compositional generalization.

Research Focus

Key Achievements

1
H-Index
1
Papers
38
Total Citations
38
Avg Citations/Paper
🏆 Most Cited Paper
The Option Keyboard: Combining Skills in Reinforcement Learning
38 citations · 2019
📈 Most Prolific Year: 2019 (1 Papers)
🤝 Key Collaborators: 10
🏛 Institutions: Google (United States)

Top Papers

  1. 1

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 11 days ago