Nanyang Associate Professor E-mail: hemant (dot) tyagi (at) ntu (dot) edu (dot) sg Office: SPMS-MAS-05-44 |
Research
interests
High-dimensional statistics, Numerical analysis, Optimization, Statistical learning theory, Inverse
problems on graphs, Learning dynamical processes
Summary of research
Broadly speaking, I am interested in inverse problems where we are given noisy and limited information about a signal of interest, and the goal is to estimate (or "learn") the unknown signal. I am particularly interested in the setting where the signal is high-dimensional, but has some intrinsic low-dimensional structure. So far, I have worked on the following problems.
- Learning high dimensional functions from point queries
- Multi-armed bandit problems
- Super-resolution of point sources
- Phase unwrapping
- Inverse problems on graphs (community detection, ranking from pairwise comparisons, group synchronization, graph matching)
- Learning dynamical systems
Job openings
Multiple PhD and postdoc positions available in 2025, see here for details.
Before emailing me, please read the job-description carefully.
Links
[Bio]
[Publications]
[Supervision]
[Teaching]
[Organization]
[Talks]