Self-Supervised Learning with Data Augmentations Provably Isolates Content from Style Julius von Kügelgen 1, 2 Y ash Sharma 3, 4 Luigi Gresele 1 Wieland Brendel
–Neural Information Processing Systems
Unlike prior work on disentanglement and independent component analysis, we allow for both nontrivial statistical and causal dependencies in the latent space.
Neural Information Processing Systems
Aug-15-2025, 17:37:40 GMT
- Country:
- South America > Chile
- North America > United States
- Virginia (0.04)
- Utah (0.04)
- Pennsylvania > Allegheny County
- Pittsburgh (0.04)
- Massachusetts > Middlesex County
- Cambridge (0.04)
- Louisiana > Orleans Parish
- New Orleans (0.04)
- Florida > Orange County
- Orlando (0.04)
- California > Santa Clara County
- Palo Alto (0.04)
- Europe
- United Kingdom > England
- Cambridgeshire > Cambridge (0.14)
- Netherlands > North Holland
- Amsterdam (0.04)
- Germany > Baden-Württemberg
- Tübingen Region > Tübingen (0.04)
- United Kingdom > England
- Asia > Japan
- Honshū > Tōhoku > Iwate Prefecture > Morioka (0.04)
- Africa > Ethiopia
- Addis Ababa > Addis Ababa (0.04)
- Genre:
- Research Report > New Finding (1.00)
- Technology: