Model Zoo: A Growing "Brain" That Learns Continually

6 Jun 2021  ·  Rahul Ramesh, Pratik Chaudhari ·

This paper argues that continual learning methods can benefit by splitting the capacity of the learner across multiple models. We use statistical learning theory and experimental analysis to show how multiple tasks can interact with each other in a non-trivial fashion when a single model is trained on them. The generalization error on a particular task can improve when it is trained with synergistic tasks, but can also deteriorate when trained with competing tasks. This theory motivates our method named Model Zoo which, inspired from the boosting literature, grows an ensemble of small models, each of which is trained during one episode of continual learning. We demonstrate that Model Zoo obtains large gains in accuracy on a variety of continual learning benchmark problems. Code is available at

PDF Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Continual Learning Cifar100 (20 tasks) Model Zoo-Continual Average Accuracy 94.99 # 1
Continual Learning Coarse-CIFAR100 Model Zoo-Continual Average Accuracy 84.27 # 1
Continual Learning Permuted MNIST Model Zoo-Continual Average Accuracy 97.71 # 2
Continual Learning Rotated MNIST Model Zoo-Continual Average Accuracy 99.66 # 1


No methods listed for this paper. Add relevant methods here