Craig Innes
Papers
1
Total Citations
15
H-Index
1
About
Craig Innes is an emerging researcher specializing in **reinforcement learning**, **inverse reward learning**, and **robotics**, with a particular focus on bridging the gap between human demonstration and autonomous agent behavior. His most notable work, "Learning Rewards from Exploratory Demonstrations Using Probabilistic Temporal Ranking" (2023), addresses a fundamental challenge in robotics: enabling systems to infer task goals without explicit cost functions or predefined goal states. By tackling this inverse problem, Innes contributes to making robotic systems more adaptable and intuitive to program through natural human interaction rather than hand-crafted reward engineering. His research sits at the intersection of probabilistic reasoning, visual servoing, and active viewpoint selection, pushing the boundaries of how autonomous agents can learn from minimal human input. The accumulation of 15 citations on a 2023 publication reflects meaningful early-stage recognition within the robotics and machine learning communities, suggesting his work addresses timely and practically relevant problems. For students and researchers working in imitation learning, human-robot interaction, or autonomous navigation, Innes represents a promising voice advancing the state of the art in data-efficient, demonstration-driven learning systems.
Research Focus
Key Achievements
Top Papers
- 1