Online Fast Adaptation and Knowledge Accumulation: a New Approach to Continual Learning

by   Massimo Caccia, et al.

Learning from non-stationary data remains a great challenge for machine learning. Continual learning addresses this problem in scenarios where the learning agent faces a stream of changing tasks. In these scenarios, the agent is expected to retain its highest performance on previous tasks without revisiting them while adapting well to the new tasks. Two new recent continual-learning scenarios have been proposed. In meta-continual learning, the model is pre-trained to minimize catastrophic forgetting when trained on a sequence of tasks. In continual-meta learning, the goal is faster remembering, i.e., focusing on how quickly the agent recovers performance rather than measuring the agent's performance without any adaptation. Both scenarios have the potential to propel the field forward. Yet in their original formulations, they each have limitations. As a remedy, we propose a more general scenario where an agent must quickly solve (new) out-of-distribution tasks, while also requiring fast remembering. We show that current continual learning, meta learning, meta-continual learning, and continual-meta learning techniques fail in this new scenario. Accordingly, we propose a strong baseline: Continual-MAML, an online extension of the popular MAML algorithm. In our empirical experiments, we show that our method is better suited to the new scenario than the methodologies mentioned above, as well as standard continual learning and meta learning approaches.


Task Agnostic Continual Learning via Meta Learning

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

Learning to Continually Learn

Continual lifelong learning requires an agent or model to learn many seq...

Reproducibility Report: La-MAML: Look-ahead Meta Learning for Continual Learning

The Continual Learning (CL) problem involves performing well on a sequen...

Meta-Learning Representations for Continual Learning

A continual learning agent should be able to build on top of existing kn...

Meta Continual Learning via Dynamic Programming

Meta-continual learning algorithms seek to rapidly train a model when fa...

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...

Value-based Bayesian Meta-reinforcement Learning and Traffic Signal Control

Reinforcement learning methods for traffic signal control has gained inc...

Please sign up or login with your details

Forgot password? Click here to reset