Robert Loftin
Papers
5
Total Citations
256
H-Index
5
About
Robert Loftin is a leading researcher in interactive machine learning, specializing in how robots and autonomous agents can learn from non-expert human teachers. His work focuses on the critical challenge of enabling people to train AI systems through natural, intuitive interactions—such as evaluative feedback, demonstrations, and even implicit cues—without requiring programming expertise. Loftin’s most influential contribution is his 2017 paper, "Interactive Learning from Policy-Dependent Human Feedback" (108 citations), which fundamentally challenged prior assumptions by showing that human feedback is not static but depends on the agent’s current behavior, leading to more robust learning algorithms. He also pioneered the concept of Cyber-Enhanced Working Dogs for search and rescue (78 citations), integrating canine intelligence with wearable sensors and AI to create a novel human-animal-robot team. His research on adapting agent action speed to improve learning (33 citations) and leveraging implicit human feedback strategies (25 citations) has advanced the practical deployment of interactive reinforcement learning. Loftin’s work bridges human-robot interaction and reinforcement learning, making him a key figure in creating AI systems that ordinary people can train effectively.
Research Focus
Key Achievements
Top Papers
- 1Interactive Learning from Policy-Dependent Human Feedback108 citations · 2017
- 2Toward Cyber-Enhanced Working Dogs for Search and Rescue78 citations · 2014
- 3
- 4
- 5Training an Agent to Ground Commands with Reward and Punishment12 citations · 2014