ALICE: Towards Understanding Adversarial Learning for Joint Distribution Matching
arXiv:1709.01215
Abstract
We investigate the non-identifiability issues associated with bidirectional adversarial training for joint distribution matching. Within a framework of conditional entropy, we propose both adversarial and non-adversarial approaches to learn desirable matched joint distributions for unsupervised and supervised tasks. We unify a broad family of adversarial models as joint distribution matching problems. Our approach stabilizes learning of unsupervised bidirectional adversarial learning methods. Further, we introduce an extension for semi-supervised learning tasks. Theoretical results are validated in synthetic data and real-world applications.
NIPS 2017 (22 pages); short version (9 pages): http://people.duke.edu/~cl319/doc/papers/nips_2017_alice.pdf
Cited by in corpus (6)
- Learning Disentangled Representations in the Imaging Domain
- PSIGAN: Joint probabilistic segmentation and image distribution matching for unpaired cross-modality adaptation based MRI segmentation
- DataLens: Scalable Privacy Preserving Training via Gradient Compression and Aggregation
- Training Discriminative Models to Evaluate Generative Ones
- Adversarial Learned Molecular Graph Inference and Generation
- Constraint-Based Visual Generation