DIFFERENCE-SEEKING GENERATIVE ADVERSARIAL NETWORK-UNSEEN SAMPLE GENERATION
Journal
8th International Conference on Learning Representations, ICLR 2020
Date Issued
2020-01-01
Author(s)
Abstract
Unseen data, which are not samples from the distribution of training data and are difficult to collect, have exhibited importance in numerous applications, (e.g., novelty detection, semi-supervised learning, and adversarial training). In this paper, we introduce a general framework called difference-seeking generative adversarial network (DSGAN), to generate various types of unseen data. Its novelty is the consideration of the probability density of the unseen data distribution as the difference between two distributions pd¯ and pd whose samples are relatively easy to collect. The DSGAN can learn the target distribution, pt, (or the unseen data distribution) from only the samples from the two distributions, pd and pd¯. In our scenario, pd is the distribution of the seen data, and pd¯ can be obtained from pd via simple operations, so that we only need the samples of pd during the training. Two key applications, semi-supervised learning and novelty detection, are taken as case studies to illustrate that the DSGAN enables the production of various unseen data. We also provide theoretical analyses about the convergence of the DSGAN.
Type
conference paper