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.
Our algorithm improves one-shot accuracy on ImageNet from 87. 6% to 93. 2% and from 88. 0% to 93. 8% on Omniglot compared to competing approaches.
#6 best model for Few-Shot Image Classification on OMNIGLOT - 1-Shot Learning
To address this paradigm, we propose novel extensions of Prototypical Networks (Snell et al., 2017) that are augmented with the ability to use unlabeled examples when producing prototypes.
The model, based on image conditional Generative Adversarial Networks, takes data from a source domain and learns to take any data item and generalise it to generate other within-class data items.
The key insight is that, in addition to features, we can transfer similarity information and this is sufficient to learn a similarity function and clustering network to perform both domain adaptation and cross-task transfer learning.
The TargetNet module is a neural network for solving a specific task and the MetaNet module aims at learning to generate functional weights for TargetNet by observing training samples.
Experiments on the MNIST and OMNIGLOT datasets show that these relaxations outperform previous discrete VAEs with Boltzmann priors.