Task Agnostic Continual Learning via Meta Learning

Abstract

While neural networks are powerful function approximators, they suffer from catastrophic forgetting when the data distribution is not stationary. One particular formalism that studies learning under non-stationary distribution is provided by continual learning, where the non-stationarity is imposed by a sequence of distinct tasks. Most methods in this space assume, however, the knowledge of task boundaries, and focus on alleviating catastrophic forgetting. In this work, we depart from this view and move the focus towards faster remembering -- i.e measuring how quickly the network recovers performance rather than measuring the network's performance without any adaptation. We argue that in many settings this can be more effective and that it opens the door to combining meta-learning and continual learning techniques, leveraging their complementary advantages. We propose a framework specific for the scenario where no information about task boundaries or task identity is given. It relies on a separation of concerns into what task is being solved and how the task should be solved. This framework is implemented by differentiating task specific parameters from task agnostic parameters, where the latter are optimized in a continual meta learning fashion, without access to multiple tasks at the same time. We showcase this framework in a supervised learning scenario and discuss the implication of the proposed formalism.

References in corpus (11)

• Adam: A Method for Stochastic Optimization
• Overcoming catastrophic forgetting in neural networks
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
• Continual Learning Through Synaptic Intelligence
• An Empirical Investigation of Catastrophic Forgetting in Gradient-Based Neural Networks
• Variational Continual Learning
• Alleviating catastrophic forgetting using context-dependent gating and synaptic stabilization
• Continual Learning via Neural Pruning
• Task Agnostic Continual Learning Using Online Variational Bayes

Cited by in corpus (24)

• Continual Learning for Recurrent Neural Networks: an Empirical Evaluation
• Understanding the Role of Training Regimes in Continual Learning
• Exploring the landscapes of "computing": digital, neuromorphic, unconventional -- and beyond
• DualNet: Continual Learning, Fast and Slow
• Orthogonal Gradient Descent for Continual Learning
• Continual Learning with Adaptive Weights (CLAW)
• Task Agnostic Continual Learning Using Online Variational Bayes with Fixed-Point Updates
• Continual Learning in Recurrent Neural Networks
• Linear Mode Connectivity in Multitask and Continual Learning
• Encoders and Ensembles for Task-Free Continual Learning
• Dimensions of Timescales in Neuromorphic Computing Systems
• Continuous Meta-Learning without Tasks
• Meta-learnt priors slow down catastrophic forgetting in neural networks
• ARCADe: A Rapid Continual Anomaly Detector
• Memory-Efficient Semi-Supervised Continual Learning: The World is its Own Replay Buffer
• Continual Representation Learning for Biometric Identification
• Understanding Continual Learning Settings with Data Distribution Drift Analysis
• Towards mental time travel: a hierarchical memory for reinforcement learning agents
• Posterior Meta-Replay for Continual Learning
• Optimizing Reusable Knowledge for Continual Learning via Metalearning
• Reviewing continual learning from the perspective of human-level intelligence
• Task-agnostic Continual Learning with Hybrid Probabilistic Models
• Subjectivity Learning Theory towards Artificial General Intelligence
• Variable-Shot Adaptation for Online Meta-Learning