On the Implicit Assumptions of GANs
This work highlights critical problems in GANs that affect research on probabilistic models, but it is incremental as it builds on prior insights.
The paper addresses issues in generative adversarial nets (GANs) that contradict theoretical guarantees due to implicit assumptions, and proposes an alternative method to sidestep these problems.
Generative adversarial nets (GANs) have generated a lot of excitement. Despite their popularity, they exhibit a number of well-documented issues in practice, which apparently contradict theoretical guarantees. A number of enlightening papers have pointed out that these issues arise from unjustified assumptions that are commonly made, but the message seems to have been lost amid the optimism of recent years. We believe the identified problems deserve more attention, and highlight the implications on both the properties of GANs and the trajectory of research on probabilistic models. We recently proposed an alternative method that sidesteps these problems.