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

Abstract

Continual learning studies agents that learn from streams of tasks without forgetting previous ones while adapting to new ones. Two recent continual-learning scenarios have opened new avenues of research. In meta-continual learning, the model is pre-trained to minimize catastrophic forgetting of previous tasks. In continual-meta learning, the aim is to train agents for faster remembering of previous tasks through adaptation. In their original formulations, both methods have limitations. We stand on their shoulders to propose a more general scenario, OSAKA, 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. We propose Continual-MAML, an online extension of the popular MAML algorithm as a strong baseline for this scenario. We empirically show that Continual-MAML is better suited to the new scenario than the aforementioned methodologies, as well as standard continual learning and meta-learning approaches.

References in corpus (18)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Overcoming catastrophic forgetting in neural networks
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Prototypical Networks for Few-shot Learning
• Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence
• An Empirical Investigation of Catastrophic Forgetting in Gradient-Based Neural Networks
• DeepStack: Expert-Level Artificial Intelligence in No-Limit Poker
• Efficient Lifelong Learning with A-GEM
• Three scenarios for continual learning
• Session-based Social Recommendation via Dynamic Graph Attention Networks
• Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning
• Reinforced Continual Learning
• Towards Robust Evaluations of Continual Learning
• Online Meta-Learning
• Learning to Continually Learn
• Defining Benchmarks for Continual Few-Shot Learning
• Learning from the Past: Continual Meta-Learning via Bayesian Graph Modeling

Cited by in corpus (6)

• Continual Learning for Recurrent Neural Networks: an Empirical Evaluation
• Wandering Within a World: Online Contextualized Few-Shot Learning
• A Policy Gradient Algorithm for Learning to Learn in Multiagent Reinforcement Learning
• Graceful Degradation and Related Fields
• Meta Continual Learning via Dynamic Programming
• Few-Shot Unsupervised Continual Learning through Meta-Examples