Three scenarios for continual learning (2019-04-15T00:00:00.000000Z)

TL;DR

Three continual learning scenarios are described based on whether at test time task identity is provided and--in case it is not--whether it must be inferred, and it is found that regularization-based approaches fail and that replaying representations of previous experiences seems required for solving this scenario.

Abstract

Standard artificial neural networks suffer from the well-known issue of catastrophic forgetting, making continual or lifelong learning difficult for machine learning. In recent years, numerous methods have been proposed for continual learning, but due to differences in evaluation protocols it is difficult to directly compare their performance. To enable more structured comparisons, we describe three continual learning scenarios based on whether at test time task identity is provided and--in case it is not--whether it must be inferred. Any sequence of well-defined tasks can be performed according to each scenario. Using the split and permuted MNIST task protocols, for each scenario we carry out an extensive comparison of recently proposed continual learning methods. We demonstrate substantial differences between the three scenarios in terms of difficulty and in terms of how efficient different methods are. In particular, when task identity must be inferred (i.e., class incremental learning), we find that regularization-based approaches (e.g., elastic weight consolidation) fail and that replaying representations of previous experiences seems required for solving this scenario.

Authors

A. Tolias

6 papers

Gido M. van de Ven

3 papers

TL;DR

Abstract

Authors

References36 items

Incremental Learning with Unlabeled Data in the Wild

Overcoming Catastrophic Forgetting With Unlabeled Data in the Wild

Re-evaluating Continual Learning Scenarios: A Categorization and Case for Strong Baselines

Generative replay with feedback connections as a general strategy for continual learning

Towards Robust Evaluations of Continual Learning

Progress & Compress: A scalable framework for continual learning

Task Agnostic Continual Learning Using Online Variational Bayes

Continual Lifelong Learning with Neural Networks: A Review

Alleviating catastrophic forgetting using context-dependent gating and synaptic stabilization

Incremental Classifier Learning with Generative Adversarial Networks

Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence

Overcoming catastrophic forgetting with hard attention to the task

Note on the quadratic penalties in elastic weight consolidation

FearNet: Brain-Inspired Model for Incremental Learning

Variational Continual Learning

Deep Generative Dual Memory Network for Continual Learning

Measuring Catastrophic Forgetting in Neural Networks

A Strategy for an Uncompromising Incremental Learner

Continual Learning with Deep Generative Replay

Continual Learning Through Synaptic Intelligence

PathNet: Evolution Channels Gradient Descent in Super Neural Networks

Overcoming catastrophic forgetting in neural networks

iCaRL: Incremental Classifier and Representation Learning

What Learning Systems do Intelligent Agents Need? Complementary Learning Systems Theory Updated

Learning without Forgetting

Tutorial on Variational Autoencoders

Pixel Recurrent Neural Networks

Variational Inference with Normalizing Flows

Distilling the Knowledge in a Neural Network

Adam: A Method for Stochastic Optimization

New Insights and Perspectives on the Natural Gradient Method

New perspectives on the natural gradient method

An Empirical Investigation of Catastrophic Forgeting in Gradient-Based Neural Networks

Auto-Encoding Variational Bayes

Metric Learning for Large Scale Image Classification: Generalizing to New Classes at Near-Zero Cost

GENERATIVE ADVERSARIAL NETS

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names