On Translation and Reconstruction Guarantees of the Cycle-Consistent Generative Adversarial Networks
Nov 28, 2022
Speakers
Anish Chakrabarty
Speaker · 0 followers
Swagatam Das
Speaker · 0 followers
About
The task of unpaired image-to-image translation has witnessed a revolution with the introduction of the cycle-consistency loss to Generative Adversarial Networks (GANs). Numerous variants, with Cycle-Consistent Adversarial Network (CycleGAN) at their forefront, have shown remarkable empirical performance. The involvement of two unalike data spaces and the existence of multiple solution maps between them are some of the facets that make such architectures unique. In this study, we investigate the statistical properties of such unpaired data translator networks between distinct spaces, bearing the additional responsibility of cycle-consistency. In a density estimation setup, we derive sharp non-asymptotic bounds on the translation errors under suitably characterized models. This, in turn, points out sufficient regularity conditions that maps must obey to carry out successful translations. We further show that cycle-consistency is achieved as a consequence of the data being successfully generated in each space based on observations from the other. In a first-of-its-kind attempt, we also provide deterministic bounds on the cumulative reconstruction error. In the process, we establish tolerable upper bounds on the discrepancy responsible for ill-posedness in such networks.The task of unpaired image-to-image translation has witnessed a revolution with the introduction of the cycle-consistency loss to Generative Adversarial Networks (GANs). Numerous variants, with Cycle-Consistent Adversarial Network (CycleGAN) at their forefront, have shown remarkable empirical performance. The involvement of two unalike data spaces and the existence of multiple solution maps between them are some of the facets that make such architectures unique. In this study, we investigate the…
Organizer
NeurIPS 2022
Account · 953 followers
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Revisiting Robustness in Graph Machine Learning
Watch later
Lukash Gosh, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Offline RL with Closed-Form Policy Improvement Operators
Watch later
Jiachen Li, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Discovering Interpretable Structural Model Errors in Climate Models
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Boosting Out-of-distribution Detection with Typical Features
Watch later
Yao Zhu, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Exploiting Reward Shifting in Value-Based Deep RL
Watch later
Hao Sun, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Panel
Watch later
Peter Lee, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%