Sasha Salter
Papers
1
Total Citations
29
H-Index
1
About
Sasha Salter is a leading researcher in robot learning, with a focus on Learning from Demonstration (LfD) and task decomposition. Their seminal work, "TACO: Learning Task Decomposition via Temporal Alignment for Control" (2018, 29 citations), introduced a novel framework that enables robots to break down complex, real-world tasks into reusable sub-policies. By aligning temporal patterns across demonstrations, TACO allows these sub-policies to be composed flexibly within and between tasks, dramatically improving data efficiency and generalization. This contribution addresses a critical bottleneck in LfD—the need for large, task-specific datasets—by enabling robots to leverage shared structure across diverse activities. Salter’s research has been recognized for its practical impact on autonomous systems, from manufacturing to assistive robotics, and their work continues to shape how robots learn hierarchical skills from human demonstrations.
Research Focus
Key Achievements
Top Papers
- 1TACO: Learning Task Decomposition via Temporal Alignment for Control29 citations · 2018