Garrett Peake

Google (United States)

Papers

1

Total Citations

5

H-Index

1

About

Garrett Peake is a rising researcher in artificial intelligence, with a primary focus on reinforcement learning and autonomous systems. His work centers on developing methods that allow AI agents to learn and improve continuously in unstructured, real-world environments without heavy human intervention. His most-cited paper, "Demonstration-Bootstrapped Autonomous Practicing via Multi-Task Reinforcement Learning" (2023, 5 citations), introduces a novel framework that combines demonstration bootstrapping with multi-task learning, enabling robots to practice and refine skills autonomously. This contribution addresses a critical bottleneck in robotics: the reliance on extensive instrumentation or human oversight for real-world learning. By demonstrating how agents can leverage prior demonstrations to bootstrap their own practice, Peake’s research paves the way for more scalable and self-sufficient AI systems. Though early in his career, his work has already garnered attention for its practical implications in robotics and autonomous decision-making. Peake’s research is particularly relevant for students and researchers interested in bridging the gap between simulated training and real-world deployment, offering a promising path toward truly autonomous learning agents.

Research Focus

Key Achievements

1
H-Index
1
Papers
5
Total Citations
5
Avg Citations/Paper
🏆 Most Cited Paper
Demonstration-Bootstrapped Autonomous Practicing via Multi-Task Reinforcement Learning
5 citations · 2023
📈 Most Prolific Year: 2023 (1 Papers)
🤝 Key Collaborators: 5
🏛 Institutions: Google (United States)

Top Papers

  1. 1

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 11 days ago