Variational Continual Learning

Abstract

This paper develops variational continual learning (VCL), a simple but general framework for continual learning that fuses online variational inference (VI) and recent advances in Monte Carlo VI for neural networks. The framework can successfully train both deep discriminative models and deep generative models in complex continual learning settings where existing tasks evolve over time and entirely new tasks emerge. Experimental results show that VCL outperforms state-of-the-art continual learning methods on a variety of tasks, avoiding catastrophic forgetting in a fully automatic way.

Published at International Conference on Learning Representations (ICLR) 2018

References in corpus (1)

• Deep Gaussian Processes for Regression using Approximate Expectation Propagation

Cited by in corpus (89)

• A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
• A continual learning survey: Defying forgetting in classification tasks
• Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence
• Three scenarios for continual learning
• Alleviating catastrophic forgetting using context-dependent gating and synaptic stabilization
• Overcoming catastrophic forgetting with hard attention to the task
• Generative replay with feedback connections as a general strategy for continual learning
• Incremental Learning for Semantic Segmentation of Large-Scale Remote Sensing Data
• Continual Learning for Recurrent Neural Networks: an Empirical Evaluation
• Practical Deep Learning with Bayesian Principles
• Variational Federated Multi-Task Learning
• Lifelong Generative Modeling
• Understanding the Role of Training Regimes in Continual Learning
• Functional Regularisation for Continual Learning with Gaussian Processes
• Online Continual Learning with Maximally Interfered Retrieval
• Gradient based sample selection for online continual learning
• Task Agnostic Continual Learning Using Online Variational Bayes
• Task Agnostic Continual Learning via Meta Learning
• Continual Learning and Catastrophic Forgetting
• CLeaR: An Adaptive Continual Learning Framework for Regression Tasks
• Orthogonal Gradient Descent for Continual Learning
• Unified Probabilistic Deep Continual Learning through Generative Replay and Open Set Recognition
• Class-incremental Learning with Pre-allocated Fixed Classifiers
• Toward Understanding Catastrophic Forgetting in Continual Learning
• Small-GAN: Speeding Up GAN Training Using Core-sets
• Maintaining Discrimination and Fairness in Class Incremental Learning
• Improving and Understanding Variational Continual Learning
• RODEO: Replay for Online Object Detection
• Improved Schemes for Episodic Memory-based Lifelong Learning
• Continual Learning with Node-Importance based Adaptive Group Sparse Regularization
• Meta-Consolidation for Continual Learning
• SupportNet: solving catastrophic forgetting in class incremental learning with support data
• Meta Continual Learning
• Lifelong Object Detection
• Flattening Sharpness for Dynamic Gradient Projection Memory Benefits Continual Learning
• Quantune: Post-training Quantization of Convolutional Neural Networks using Extreme Gradient Boosting for Fast Deployment
• The Effectiveness of Memory Replay in Large Scale Continual Learning
• Continual State Representation Learning for Reinforcement Learning using Generative Replay
• Bayesian Variational Autoencoders for Unsupervised Out-of-Distribution Detection
• Continual Classification Learning Using Generative Models
• Selective Amnesia: On Efficient, High-Fidelity and Blind Suppression of Backdoor Effects in Trojaned Machine Learning Models
• Task Agnostic Continual Learning Using Online Variational Bayes with Fixed-Point Updates
• Class Gradient Projection For Continual Learning
• Optimization and Generalization of Regularization-Based Continual Learning: a Loss Approximation Viewpoint
• VAEM: a Deep Generative Model for Heterogeneous Mixed Type Data
• Reconciling meta-learning and continual learning with online mixtures of tasks
• Linear Mode Connectivity in Multitask and Continual Learning
• Variational Prototype Replays for Continual Learning
• Self-Supervised Learning Aided Class-Incremental Lifelong Learning
• Continual Domain-Tuning for Pretrained Language Models
• Powerpropagation: A sparsity inducing weight reparameterisation
• Continual Model-Based Reinforcement Learning with Hypernetworks
• Continual Learning Using Bayesian Neural Networks
• How to Evaluate the Next System: Automatic Dialogue Evaluation from the Perspective of Continual Learning
• Unifying Regularisation Methods for Continual Learning
• Layerwise Optimization by Gradient Decomposition for Continual Learning
• Wide Neural Networks Forget Less Catastrophically
• Continual Learning Using Multi-view Task Conditional Neural Networks
• Efficient Architecture Search for Continual Learning
• Natural continual learning: success is a journey, not (just) a destination
• Bayesian Reasoning with Trained Neural Networks
• Continual Learning via Bit-Level Information Preserving
• GAN Memory with No Forgetting
• Low Complexity Approximate Bayesian Logistic Regression for Sparse Online Learning
• Single-Net Continual Learning with Progressive Segmented Training (PST)
• Frosting Weights for Better Continual Training
• Intelligence, physics and information -- the tradeoff between accuracy and simplicity in machine learning
• Generative Memory for Lifelong Reinforcement Learning
• Graph-Based Continual Learning
• Variational Bayes: A report on approaches and applications
• Human Inspired Progressive Alignment and Comparative Learning for Grounded Word Acquisition
• Reducing Geographic Disparities in Automatic Speech Recognition via Elastic Weight Consolidation
• Contextual HyperNetworks for Novel Feature Adaptation
• Hidden Markov Neural Networks
• Deep Bayesian Unsupervised Lifelong Learning
• Tuned Compositional Feature Replays for Efficient Stream Learning
• Dynamic Continual Learning: Harnessing Parameter Uncertainty for Improved Network Adaptation
• Bilevel Continual Learning
• Shared and Private VAEs with Generative Replay for Continual Learning
• On regret bounds for continual single-index learning
• Center Loss Regularization for Continual Learning
• Continual Novelty Detection
• ZS-IL: Looking Back on Learned Experiences For Zero-Shot Incremental Learning
• Measuring Information Transfer in Neural Networks
• Continual Learning of Multi-modal Dynamics with External Memory
• Dynamic VAEs with Generative Replay for Continual Zero-shot Learning
• Class Incremental Online Streaming Learning
• Lifelong Learning Process: Self-Memory Supervising and Dynamically Growing Networks
• Schematic Memory Persistence and Transience for Efficient and Robust Continual Learning