Infinite Mixture Prototypes for Few-Shot Learning
This work addresses the challenge of few-shot learning for AI systems by introducing a more flexible representation method, though it is incremental as it builds on existing prototypical approaches.
The paper tackles the problem of few-shot learning by proposing infinite mixture prototypes to adaptively represent complex data distributions, resulting in a 25% absolute accuracy improvement over prototypical networks on alphabets and state-of-the-art semi-supervised accuracy.
We propose infinite mixture prototypes to adaptively represent both simple and complex data distributions for few-shot learning. Our infinite mixture prototypes represent each class by a set of clusters, unlike existing prototypical methods that represent each class by a single cluster. By inferring the number of clusters, infinite mixture prototypes interpolate between nearest neighbor and prototypical representations, which improves accuracy and robustness in the few-shot regime. We show the importance of adaptive capacity for capturing complex data distributions such as alphabets, with 25% absolute accuracy improvements over prototypical networks, while still maintaining or improving accuracy on the standard Omniglot and mini-ImageNet benchmarks. In clustering labeled and unlabeled data by the same clustering rule, infinite mixture prototypes achieves state-of-the-art semi-supervised accuracy. As a further capability, we show that infinite mixture prototypes can perform purely unsupervised clustering, unlike existing prototypical methods.