Importance Weighted Generative Networks
This addresses sample selection bias in generative modeling, which is an incremental improvement for machine learning applications dealing with biased data.
The paper tackles the problem of training generative networks when data is subject to selection bias or from a related distribution, by introducing importance weighting methods to estimate the loss with respect to a target distribution, resulting in theoretical guarantees and impressive empirical performance.
Deep generative networks can simulate from a complex target distribution, by minimizing a loss with respect to samples from that distribution. However, often we do not have direct access to our target distribution - our data may be subject to sample selection bias, or may be from a different but related distribution. We present methods based on importance weighting that can estimate the loss with respect to a target distribution, even if we cannot access that distribution directly, in a variety of settings. These estimators, which differentially weight the contribution of data to the loss function, offer both theoretical guarantees and impressive empirical performance.