Continual Unsupervised Representation Learning

Abstract

Continual learning aims to improve the ability of modern learning systems to deal with non-stationary distributions, typically by attempting to learn a series of tasks sequentially. Prior art in the field has largely considered supervised or reinforcement learning tasks, and often assumes full knowledge of task labels and boundaries. In this work, we propose an approach (CURL) to tackle a more general problem that we will refer to as unsupervised continual learning. The focus is on learning representations without any knowledge about task identity, and we explore scenarios when there are abrupt changes between tasks, smooth transitions from one task to another, or even when the data is shuffled. The proposed approach performs task inference directly within the model, is able to dynamically expand to capture new concepts over its lifetime, and incorporates additional rehearsal-based techniques to deal with catastrophic forgetting. We demonstrate the efficacy of CURL in an unsupervised learning setting with MNIST and Omniglot, where the lack of labels ensures no information is leaked about the task. Further, we demonstrate strong performance compared to prior art in an i.i.d setting, or when adapting the technique to supervised tasks such as incremental class learning.

NeurIPS 2019

Cited by in corpus (20)

• Lifelong Teacher-Student Network Learning
• DualNet: Continual Learning, Fast and Slow
• From Machine Learning to Robotics: Challenges and Opportunities for Embodied Intelligence
• Hypernetworks for Continual Semi-Supervised Learning
• Learning latent representations across multiple data domains using Lifelong VAEGAN
• Modelling Latent Skills for Multitask Language Generation
• Lifelong Learning using Eigentasks: Task Separation, Skill Acquisition, and Selective Transfer
• Bayesian Structure Adaptation for Continual Learning
• Continual Learning from the Perspective of Compression
• Task-agnostic Continual Learning with Hybrid Probabilistic Models
• Semi-Supervised Class Discovery
• Streaming Self-Training via Domain-Agnostic Unlabeled Images
• Towards Recognizing New Semantic Concepts in New Visual Domains
• Deep Bayesian Unsupervised Lifelong Learning
• Discriminative Distillation to Reduce Class Confusion in Continual Learning
• POLA: Online Time Series Prediction by Adaptive Learning Rates
• Lifelong Mixture of Variational Autoencoders
• Lifelong Twin Generative Adversarial Networks
• Meta-Learning with Variational Bayes
• A Conditional Generative Matching Model for Multi-lingual Reply Suggestion