Posterior Meta-Replay for Continual Learning

Christian Henning, Maria Cervera, Francesco D'Angelo, Johannes Von Oswald, Regina Traber, Benjamin Ehret, Seijin Kobayashi, Benjamin F Grewe, Joao Sacramento

Keywords: Continual Learning, Uncertainty Estimation, Bayesian Neural Networks, Task Inference, Hypernetworks

TL;DR: Learning task-specific posteriors continually in a shared meta-model and studying the properties of predictive uncertainty for task inference.

Abstract: Learning a sequence of tasks without access to i.i.d. observations is a widely studied form of continual learning (CL) that remains challenging. In principle, Bayesian learning directly applies to this setting, since recursive and one-off Bayesian updates yield the same result. In practice, however, recursive updating often leads to poor trade-off solutions across tasks because approximate inference is necessary for most models of interest. Here, we describe an alternative Bayesian approach where task-conditioned parameter distributions are continually inferred from data. We offer a practical deep learning implementation of our framework based on probabilistic task-conditioned hypernetworks, an approach we term posterior meta-replay. Experiments on standard benchmarks show that our probabilistic hypernetworks compress sequences of posterior parameter distributions with virtually no forgetting. We obtain considerable performance gains compared to existing Bayesian CL methods, and identify task inference as our major limiting factor. This limitation has several causes that are independent of the considered sequential setting, opening up new avenues for progress in CL.

Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.

Code: https://github.com/chrhenning/posterior_replay_cl

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2103.01133/code)