Continual Density Ratio Estimation (CDRE): A new method for evaluating generative models in continual learning

Yu Chen, Song Liu, Tom Diethe, Peter Flach

TL;DR: We propose a new method (Continual Density Ratio Estimation) for evaluating generative models in continual learning using f-divegences.

Abstract: We propose a new method Continual Density Ratio Estimation (CDRE), which can estimate density ratios between a target distribution of real samples and a distribution of samples generated by a model while the model is changing over time and the data of the target distribution is not available after a certain time point. This method perfectly fits the setting of continual learning, in which one model is supposed to learn different tasks sequentially and the most crucial restriction is that model has none or very limited access to the data of all learned tasks. Through CDRE, we can evaluate generative models in continual learning using f-divergences. To the best of our knowledge, there is no existing method that can evaluate generative models under the setting of continual learning without storing real samples from the target distribution.

Keywords: density ratio estimation, continual learning, evaluation, generative model, f divergence