Toward Understanding Catastrophic Forgetting in Continual Learning

Abstract

We study the relationship between catastrophic forgetting and properties of task sequences. In particular, given a sequence of tasks, we would like to understand which properties of this sequence influence the error rates of continual learning algorithms trained on the sequence. To this end, we propose a new procedure that makes use of recent developments in task space modeling as well as correlation analysis to specify and analyze the properties we are interested in. As an application, we apply our procedure to study two properties of a task sequence: (1) total complexity and (2) sequential heterogeneity. We show that error rates are strongly and positively correlated to a task sequence's total complexity for some state-of-the-art algorithms. We also show that, surprisingly, the error rates have no or even negative correlations in some cases to sequential heterogeneity. Our findings suggest directions for improving continual learning benchmarks and methods.

References in corpus (17)

• Overcoming catastrophic forgetting in neural networks
• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence
• Continual Learning Through Synaptic Intelligence
• An Empirical Investigation of Catastrophic Forgetting in Gradient-Based Neural Networks
• Variational Continual Learning
• Overcoming Catastrophic Forgetting by Incremental Moment Matching
• Online Structured Laplace Approximations For Overcoming Catastrophic Forgetting
• Towards Robust Evaluations of Continual Learning
• Towards a Neural Statistician
• Deep Generative Dual Memory Network for Continual Learning
• On Quadratic Penalties in Elastic Weight Consolidation
• Life-Long Disentangled Representation Learning with Cross-Domain Latent Homologies
• Task Agnostic Continual Learning Using Online Variational Bayes
• Improving and Understanding Variational Continual Learning
• Partitioned Variational Inference: A unified framework encompassing federated and continual learning
• Bayesian Incremental Learning for Deep Neural Networks

Cited by in corpus (20)

• A continual learning survey: Defying forgetting in classification tasks
• LEEP: A New Measure to Evaluate Transferability of Learned Representations
• A reinforcement learning approach to rare trajectory sampling
• Orthogonal Gradient Descent for Continual Learning
• Improved Schemes for Episodic Memory-based Lifelong Learning
• Anatomy of Catastrophic Forgetting: Hidden Representations and Task Semantics
• A Theoretical Analysis of Catastrophic Forgetting through the NTK Overlap Matrix
• Optimization and Generalization of Regularization-Based Continual Learning: a Loss Approximation Viewpoint
• Advances in Continual Graph Learning for Anti-Money Laundering Systems: A Comprehensive Review
• Linear Mode Connectivity in Multitask and Continual Learning
• Generalisation Guarantees for Continual Learning with Orthogonal Gradient Descent
• On Class Orderings for Incremental Learning
• Dissecting Catastrophic Forgetting in Continual Learning by Deep Visualization
• When Do Curricula Work?
• Task-similarity Aware Meta-learning through Nonparametric Kernel Regression
• Wide Neural Networks Forget Less Catastrophically
• CLOPS: Continual Learning of Physiological Signals
• Towards a practical measure of interference for reinforcement learning
• Explaining How Deep Neural Networks Forget by Deep Visualization
• Dendritic Self-Organizing Maps for Continual Learning