Sridhar Mahadevan
homepage:http://www.cs.umass.edu/~mahadeva/
search externally:   Google Scholar,   Springer,   CiteSeer,   Microsoft Academic Search,   Scirus ,   DBlife

Description

Broadly, my research interests are in artificial intelligence, specifically the credit assignment problem, and computational models of learning and sequential decision-making. I am currently investigating the problem of representation learning , which extracts regularities that apply both across domains and disparate learning modalities, including clustering, supervised learning, and reinforcement learning. I am investigating techniques for constructing new basis representations using ideas from harmonic analysis, where spatial and temporal information is remapped into a "frequency" oriented coordinate system.


Lectures:

lecture
flag Proximal Reinforcement Learning: Learning to Act in Primal Dual Spaces
as author at  2nd Multidisciplinary Conference on Reinforcement Learning and Decision Making (RLDM), Edmonton 2015,
324 views
  lecture
flag Adaptive Mesh Compression in 3D Computer Graphics using Multiscale Manifold Learning
as author at  24th Annual International Conference on Machine Learning (ICML), Corvallis 2007,
1639 views