Posterior Meta-Replay for Continual Learning

by   Christian Henning, et al.

Continual Learning (CL) algorithms have recently received a lot of attention as they attempt to overcome the need to train with an i.i.d. sample from some unknown target data distribution. Building on prior work, we study principled ways to tackle the CL problem by adopting a Bayesian perspective and focus on continually learning a task-specific posterior distribution via a shared meta-model, a task-conditioned hypernetwork. This approach, which we term Posterior-replay CL, is in sharp contrast to most Bayesian CL approaches that focus on the recursive update of a single posterior distribution. The benefits of our approach are (1) an increased flexibility to model solutions in weight space and therewith less susceptibility to task dissimilarity, (2) access to principled task-specific predictive uncertainty estimates, that can be used to infer task identity during test time and to detect task boundaries during training, and (3) the ability to revisit and update task-specific posteriors in a principled manner without requiring access to past data. The proposed framework is versatile, which we demonstrate using simple posterior approximations (such as Gaussians) as well as powerful, implicit distributions modelled via a neural network. We illustrate the conceptual advance of our framework on low-dimensional problems and show performance gains on computer vision benchmarks.


page 7

page 26

page 28

page 34

page 35


Task Agnostic Continual Learning via Meta Learning

While neural networks are powerful function approximators, they suffer f...

A Unifying Bayesian View of Continual Learning

Some machine learning applications require continual learning - where da...

Learning to modulate random weights can induce task-specific contexts for economical meta and continual learning

Neural networks are vulnerable to catastrophic forgetting when data dist...

Kernel Continual Learning

This paper introduces kernel continual learning, a simple but effective ...

ScrollNet: Dynamic Weight Importance for Continual Learning

The principle underlying most existing continual learning (CL) methods i...

Efficient Continual Adaptation for Generative Adversarial Networks

We present a continual learning approach for generative adversarial netw...

Please sign up or login with your details

Forgot password? Click here to reset