Disentangling Visual Transformers: Patch-level Interpretability for Image Classification
Visual transformers have achieved remarkable performance in image classification tasks, but this performance gain has come at the cost of interpretability. One of the main obstacles to the interpretation of transformers is the self-attention mechanism, which mixes visual information across the whole image in a complex way. In this paper, we propose Hindered Transformer (HiT), a novel interpretable by design architecture inspired by visual transformers. Our proposed architecture rethinks the design of transformers to better disentangle patch influences at the classification stage. Ultimately, HiT can be interpreted as a linear combination of patch-level information. We show that the advantages of our approach in terms of explicability come with a reasonable trade-off in performance, making it an attractive alternative for applications where interpretability is paramount.
Code (0)
등록된 구현이 없습니다.
Tasks
image-classificationImage ClassificationMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Patch-level Representation Learning for Self-supervised Vision Transformers
Recent self-supervised learning (SSL) methods have shown impressive results in learning visual representations from unlabeled images. This paper aims to improve their performance further by utilizing the architectural ad…
Instance Segmentationobject-detectionObject DetectionRepresentation Learning+3ViT-NeT: Interpretable Vision Transformers with Neural Tree Decoder
Vision transformers (ViTs), which have demonstrated a state-of-the-art performance in image classification, can also visualize global interpretations through attention-based contributions. How- ever, the complexity of th…
Decision MakingDecoderFine-Grained Image ClassificationFine-Grained Visual Categorization+2IA-RED$^2$: Interpretability-Aware Redundancy Reduction for Vision Transformers
The self-attention-based model, transformer, is recently becoming the leading backbone in the field of computer vision. In spite of the impressive success made by transformers in a variety of vision tasks, it still suffe…
Efficient ViTsBeyond Semantics: Disentangling Information Scope in Sparse Autoencoders for CLIP
Sparse Autoencoders (SAEs) have emerged as a powerful tool for interpreting the internal representations of CLIP vision encoders, yet existing analyses largely focus on the semantic meaning of individual features. We int…
Visual Interpretability for Deep Learning: a Survey
This paper reviews recent studies in understanding neural-network representations and learning neural networks with interpretable/disentangled middle-layer representations. Although deep neural networks have exhibited su…
Deep LearningExplainable artificial intelligenceSurvey