Few-shot Open-set Recognition using Meta-learning

Hao Kang2

Haoxiang Li2

Gang Hua2


UC San Diego1, Wormpex AI Research2

Overview


The problem of open-set recognition is considered. While previous approaches only consider this problem in the context of large-scale classifier training, we seek a unified solution for this and the low-shot classification setting. It is argued that the classic softmax classifier is a poor solution for open-set recognition, since it tends to overfit on the training classes. Randomization is then proposed as a solution to this problem. This suggests the use of meta-learning techniques, commonly used for few-shot classification, for the solution of open-set recognition. A new oPen sEt mEta LEaRning (PEELER) algorithm is then introduced. This combines the random selection of a set of novel classes per episode, a loss that maximizes the posterior entropy for examples of those classes, and a new metric learning formulation based on the Mahalanobis distance. Experimental results show that PEELER achieves state of the art open set recognition performance for both few-shot and large-scale recognition. On CIFAR and miniImageNet, it achieves substantial gains in seen/unseen class detection AUROC for a given seen-class classification accuracy.

paper

Published in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020.

Arxiv

Repository

Bibtex

Models


paper

Architecture: Schematic representation of oPen sEt mEta LEaRning algorithm.

Highlights

  1. Open-set sampling: sample images from unseen clases in meta-testing set.
  2. Open-set training: maximizing the posterior entorpy of unseen sammples.
  3. paper
    paper
  4. Gaussian embedding: implement Mahalanobis distance in feature space
  5. paper

Code

Training, evaluation and deployment code available on GitHub.

Video


Authors



Bo Liu

UC San Diego
haokang

Hao Kang

Wormpex AI Research
haoxiangli

Haoxiang Li

Wormpex AI Research
ganghua

Gang Hua

Wormpex AI Research

Acknowledgements

Gang Hua was supported partly by National Key R&D Program of China Grant 2018AAA0101400 and NSFC Grant 61629301. Bo Liu and Nuno Vasconcelos were partially supported by NSF awards IIS-1637941, IIS-1924937, and NVIDIA GPU donations.