Continual Learning with Adaptive Weights (CLAW)
arXiv:1911.09514
Abstract
Approaches to continual learning aim to successfully learn a set of related tasks that arrive in an online manner. Recently, several frameworks have been developed which enable deep learning to be deployed in this learning scenario. A key modelling decision is to what extent the architecture should be shared across tasks. On the one hand, separately modelling each task avoids catastrophic forgetting but it does not support transfer learning and leads to large models. On the other hand, rigidly specifying a shared component and a task-specific part enables task transfer and limits the model size, but it is vulnerable to catastrophic forgetting and restricts the form of task-transfer that can occur. Ideally, the network should adaptively identify which parts of the network to share in a data driven way. Here we introduce such an approach called Continual Learning with Adaptive Weights (CLAW), which is based on probabilistic modelling and variational inference. Experiments show that CLAW achieves state-of-the-art performance on six benchmarks in terms of overall continual learning performance, as measured by classification accuracy, and in terms of addressing catastrophic forgetting.
ICLR 2020
References in corpus (38)
- Overcoming catastrophic forgetting in neural networks
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
- Semi-Supervised Learning with Deep Generative Models
- NIPS 2016 Tutorial: Generative Adversarial Networks
- Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence
- Continual Learning Through Synaptic Intelligence
- An Empirical Investigation of Catastrophic Forgetting in Gradient-Based Neural Networks
- PathNet: Evolution Channels Gradient Descent in Super Neural Networks
- Experience Replay for Continual Learning
- Bayesian Compression for Deep Learning
- Variational Dropout Sparsifies Deep Neural Networks
- FearNet: Brain-Inspired Model for Incremental Learning
- Progress & Compress: A scalable framework for continual learning
- Measuring Catastrophic Forgetting in Neural Networks
- Generative replay with feedback connections as a general strategy for continual learning
- Less-forgetting Learning in Deep Neural Networks
- Deep Generative Dual Memory Network for Continual Learning
- BERT and PALs: Projected Attention Layers for Efficient Adaptation in Multi-Task Learning
- Functional Regularisation for Continual Learning with Gaussian Processes
- A comprehensive, application-oriented study of catastrophic forgetting in DNNs
- Selfless Sequential Learning
- Task Agnostic Continual Learning Using Online Variational Bayes
- Task Agnostic Continual Learning via Meta Learning
- Meta Continual Learning
- Online Contrastive Divergence with Generative Replay: Experience Replay without Storing Data
- Scalable and Order-robust Continual Learning with Additive Parameter Decomposition
- Keep and Learn: Continual Learning by Constraining the Latent Space for Knowledge Preservation in Neural Networks
- Incremental Learning with Maximum Entropy Regularization: Rethinking Forgetting and Intransigence
- Learning Neural Networks with Adaptive Regularization
- Autoencoder-Based Incremental Class Learning without Retraining on Old Data
- Efficient Multitask Feature and Relationship Learning
- Continual Learning in Deep Neural Network by Using a Kalman Optimiser
- Bayesian Optimized Continual Learning with Attention Mechanism
- Conditional Computation for Continual Learning
- Continual Learning via Online Leverage Score Sampling
- One Big Net For Everything
- Single-Net Continual Learning with Progressive Segmented Training (PST)