Seminar, Hosted by European Association for Signal Processing
Task-Informed and Controllable Generative Flows for Inverse Problems
Rice University
Abstract
How can a generative flow use observations and structural constraints when solving an inverse problem? Segarra describes observation-conditioned initialization for graph reconstruction, adaptive multipliers that enforce graph constraints during sampling, and a denoiser that learns finite-time flow maps. A lookahead parameter changes the balance between a conditional-mean estimate and a perceptually plausible sample.
Topics
- inverse problems
- graph reconstruction, broader topic: reconstruction
- Generative Flows, search World Wide
- observation-conditioned initialization, search World Wide
- structural constraints
- adaptive multipliers, search World Wide
- sampling
- denoiser, search World Wide
- finite-time flow maps, search World Wide
- lookahead parameter, search World Wide