Jeffrey Mark Siskind
Papers
7
Total Citations
55
H-Index
4
About
Jeffrey Mark Siskind is a pioneering researcher at the intersection of computer vision, robotics, and natural language processing, with a core focus on grounding language in physical action. His most influential work, "Learning physically-instantiated game play through visual observation" (23 citations), introduced an integrated vision and robotic system capable of learning and playing physical board games like TIC TAC TOE by watching human demonstrations—a landmark step toward machines that learn from visual experience. Siskind has made major contributions to goal-directed reinforcement learning, notably through his Floyd-Warshall Reinforcement Learning framework (10 citations), which enables agents to leverage past experiences to reach new goals in static environments. He is also recognized for his unified framework for grounding natural-language semantics in robotic driving, as demonstrated in "Driving Under the Influence (of Language)" (8 citations) and subsequent work on dialogue-driven navigation in unknown indoor environments (3 citations). His research uniquely bridges language acquisition, visual reasoning, and physical interaction, advancing the frontier of embodied AI. Siskind’s work on "Seeing Unseeability to See the Unseeable" (3 citations) further showcases his innovative approach to inferring occluded structures from visual evidence.
Research Focus
Key Achievements
Top Papers
- 1Learning physically-instantiated game play through visual observation23 citations · 2010
- 2
- 3Driving Under the Influence (of Language)8 citations · 2017
- 4
- 5
- 6Seeing Unseeability to See the Unseeable3 citations · 2012
- 7Robot Language Learning, Generation, and Comprehension3 citations · 2015