MTUNet: Few-shot Image Classification with Visual Explanations
Bowen Wang, Liangzhi Li, Manisha Verma, Yuta Nakashima, Ryo Kawasaki, Hajime Nagahara
Abstract
Few-shot learning (FSL) approaches, mostly neural network-based, are assuming that the pre-trained knowledge can be obtained from base (seen) categories and transferred to novel (unseen) categories. However, the black-box nature of neural networks makes it difficult to understand what is actually transferred, which may hamper its application in some risk-sensitive areas. In this paper, we reveal a new way to perform explainable FSL for image classification, using discriminative patterns and pairwise matching. Experimental results prove that the proposed method can achieve satisfactory explainability on two mainstream datasets. Code is available*.
Topics & Concepts
Discriminative modelComputer scienceArtificial intelligenceContextual image classificationPairwise comparisonMatching (statistics)Artificial neural networkImage (mathematics)Code (set theory)Shot (pellet)Pattern recognition (psychology)Machine learningBlack boxOne shotMathematicsEngineeringOrganic chemistryMechanical engineeringStatisticsChemistrySet (abstract data type)Programming languageDomain Adaptation and Few-Shot LearningMultimodal Machine Learning ApplicationsAdversarial Robustness in Machine Learning