Michael Laskin
Papers
3
Total Citations
34
H-Index
3
About
Michael Laskin is a leading researcher at the intersection of reinforcement learning, robotics, and unsupervised representation learning. His work focuses on enabling data-efficient, real-world robotic learning by bridging the gap between simulated environments and physical systems. Laskin’s most impactful contributions include pioneering contrastive pre-training and data augmentation techniques for visual robotic control, which dramatically improve the sample efficiency of reinforcement learning policies in real-robot settings—a breakthrough that has garnered over 27 citations across his key papers. He also introduced "Skill Preferences," a novel framework that extracts and executes robotic skills from human feedback, addressing the challenge of learning long-horizon tasks from biased offline demonstration datasets. This work, cited 7 times, offers a path toward more robust and usable skill acquisition. Laskin’s research is notable for its practical focus on making robot learning faster and more reliable, directly impacting how autonomous systems can be trained with limited real-world data. His achievements mark him as a rising figure in robotic learning and AI.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3