Learning Independent Features with Adversarial Nets for Non-linear ICA
This addresses the challenge of estimating mutual information in ICA for researchers in machine learning, though it appears incremental as it builds on existing adversarial methods for dependence measures.
The paper tackles the problem of learning independent features for tasks like source separation by proposing adversarial objectives that implicitly optimize statistical dependence measures without computing probability densities, and experiments show the method works for both linear and non-linear ICA across various model architectures.
Reliable measures of statistical dependence could be useful tools for learning independent features and performing tasks like source separation using Independent Component Analysis (ICA). Unfortunately, many of such measures, like the mutual information, are hard to estimate and optimize directly. We propose to learn independent features with adversarial objectives which optimize such measures implicitly. These objectives compare samples from the joint distribution and the product of the marginals without the need to compute any probability densities. We also propose two methods for obtaining samples from the product of the marginals using either a simple resampling trick or a separate parametric distribution. Our experiments show that this strategy can easily be applied to different types of model architectures and solve both linear and non-linear ICA problems.