Meta Continual Learning via Dynamic Programming
arXiv:2008.02219
Abstract
Meta continual learning algorithms seek to train a model when faced with similar tasks observed in a sequential manner. Despite promising methodological advancements, there is a lack of theoretical frameworks that enable analysis of learning challenges such as generalization and catastrophic forgetting. To that end, we develop a new theoretical approach for meta continual learning~(MCL) where we mathematically model the learning dynamics using dynamic programming, and we establish conditions of optimality for the MCL problem. Moreover, using the theoretical framework, we derive a new dynamic-programming-based MCL method that adopts stochastic-gradient-driven alternating optimization to balance generalization and catastrophic forgetting. We show that, on MCL benchmark data sets, our theoretically grounded method achieves accuracy better than or comparable to that of existing state-of-the-art methods.
References in corpus (9)
- Overcoming catastrophic forgetting in neural networks
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- On the difficulty of training Recurrent Neural Networks
- Continual Learning Through Synaptic Intelligence
- On Tiny Episodic Memories in Continual Learning
- Lifelong Learning with Dynamically Expandable Networks
- Learning to Continually Learn
- Meta-Learning with Warped Gradient Descent
- Online Fast Adaptation and Knowledge Accumulation: a New Approach to Continual Learning