Sridhar Mahadevan
Sridhar Mahadevan’s early work focused on learning apprentices and explanation-based learning. The program asked how a system could acquire useful structure from experience and explanations. His work progressed from reinforcement learning, hierarchical decisions and robot learning to spectral methods and proto-value functions. These methods address representation discovery for learning and decision problems. Mahadevan’s later work covers category theory, causal inference and compositionality in AI. His books address categorical intelligence and learning from failures of compositional structure.
- Works
- 201
- Citations
- 7,658
- h-index
- 39
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
1980s–1990s
Learning with explanations
Sridhar Mahadevan’s early work focused on learning apprentices and explanation-based learning. The program asked how a system could acquire useful structure from experience and explanations.
-
1990s–2010s
Agents and representations
His work progressed from reinforcement learning, hierarchical decisions and robot learning to spectral methods and proto-value functions. These methods address representation discovery for learning and decision problems.
-
2020s
Compositional learning
Mahadevan’s later work covers category theory, causal inference and compositionality in AI. His books address categorical intelligence and learning from failures of compositional structure.
Sources and credits
Photo: people.cs.umass.edu; creator not stated. License: Not stated by source. Source: people.cs.umass.edu. Cropped to a square and resized. Ask us to remove or replace this photo.