Continual Learning with Deep Generative Replay (2017-05-01T00:00:00.000000Z)

TL;DR

The Deep Generative Replay is proposed, a novel framework with a cooperative dual model architecture consisting of a deep generative model ("generator") and a task solving model ("solver"), with only these two models, training data for previous tasks can easily be sampled and interleaved with those for a new task.

Abstract

Attempts to train a comprehensive artificial intelligence capable of solving multiple tasks have been impeded by a chronic problem called catastrophic forgetting. Although simply replaying all previous data alleviates the problem, it requires large memory and even worse, often infeasible in real world applications where the access to past data is limited. Inspired by the generative nature of hippocampus as a short-term memory system in primate brain, we propose the Deep Generative Replay, a novel framework with a cooperative dual model architecture consisting of a deep generative model ("generator") and a task solving model ("solver"). With only these two models, training data for previous tasks can easily be sampled and interleaved with those for a new task. We test our methods in several sequential learning settings involving image classification tasks.

Authors

Jiwon Kim

2 papers

Jung Kwon Lee

3 papers

Hanul Shin

2 papers

TL;DR

Abstract

Authors

References36 items

Improved Training of Wasserstein GANs

Overcoming Catastrophic Forgetting by Incremental Moment Matching

NIPS 2016 Tutorial: Generative Adversarial Networks

Overcoming catastrophic forgetting in neural networks

Online Contrastive Divergence with Generative Replay: Experience Replay without Storing Data

Learning without Forgetting

Human-level control through deep reinforcement learning

A biologically inspired dual-network memory model for reduction of catastrophic forgetting

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

Auto-Encoding Variational Bayes

Compete to Compute

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation

Creating a False Memory in the Hippocampus

Play it again: reactivation of waking experience and memory

The development of object categorization in young children: hierarchical inclusiveness, age, perceptual attribute, and group versus individual analyses.

Deep Boltzmann Machines

Internally Generated Reactivation of Single Neurons in Human Hippocampus During Free Recall

Sleep-dependent memory consolidation and reconsolidation.

Evidence for large long-term memory capacities in baboons and pigeons and its implications for learning and the evolution of cognition

Memory retention – the synaptic stability versus plasticity dilemma

Long-term dendritic spine stability in the adult cortex

Multiple Classifier Systems

Catastrophic forgetting in connectionist networks

Catastrophic Forgetting, Rehearsal and Pseudorehearsal

Catastrophic forgetting in neural networks: the role of rehearsal mechanisms

Infants' ability to draw inferences about nonobvious object properties: evidence from exploratory play.

Connectionist models of recognition memory: constraints imposed by learning and forgetting functions.

GENERATIVE ADVERSARIAL NETS

Dropout: a simple way to prevent neural networks from overfitting

Reading Digits in Natural Images with Unsupervised Feature Learning

Continual Learning

Avoiding catastrophic forgetting by coupling two reverberating neural networks

Gradient-based learning applied to document recognition

Catastrophic Interference in Connectionist Networks: The Sequential Learning Problem

Using fast weights to deblur old memories

Opinion TRENDS in Cognitive Sciences Vol.6 No.12 December 2002

Field of Study

Venue Information

Name

Type

URL

Alternate Names