paper-with-me

홈 › Papers

Interpretable Compositional Convolutional Neural Networks

2021-07-09 · Wen Shen, Zhihua Wei, Shikun Huang, BinBin Zhang, Jiaqi Fan, Ping Zhao, Quanshi Zhang

The reasonable definition of semantic interpretability presents the core challenge in explainable AI. This paper proposes a method to modify a traditional convolutional neural network (CNN) into an interpretable compositional CNN, in order to learn filters that encode meaningful visual patterns in intermediate convolutional layers. In a compositional CNN, each filter is supposed to consistently represent a specific compositional object part or image region with a clear meaning. The compositional CNN learns from image labels for classification without any annotations of parts or regions for supervision. Our method can be broadly applied to different types of CNNs. Experiments have demonstrated the effectiveness of our method.

📄 PDF Abstract BibTeX arXiv:2107.04474

Code (1)

ada-shen/icCNN 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Compositional Convolutional Neural Networks: A Robust and Interpretable Model for Object Recognition under Occlusion

2020-06-28 · Adam Kortylewski, Qing Liu, Angtian Wang, Yihong Sun 외

Computer vision systems in real-world applications need to be robust to partial occlusion while also being explainable. In this work, we show that black-box deep convolutional neural networks (DCNNs) have only limited ro…

image-classificationImage Classificationobject-detectionObject Detection+1

Bridging Continuous and Discrete Spaces: Interpretable Sentence Representation Learning via Compositional Operations

2023-05-24 · James Y. Huang, Wenlin Yao, Kaiqiang Song, Hongming Zhang 외

Traditional sentence embedding models encode sentences into vector representations to capture useful properties such as the semantic similarity between sentences. However, in addition to similarity, sentence semantics ca…

DecoderRepresentation LearningSemantic SimilaritySemantic Textual Similarity+5

Inducing Hierarchical Compositional Model by Sparsifying Generator Network

2019-09-10 · CVPR 2020 6 · Xianglei Xing, Tianfu Wu, Song-Chun Zhu, Ying Nian Wu

This paper proposes to learn hierarchical compositional AND-OR model for interpretable image synthesis by sparsifying the generator network. The proposed method adopts the scene-objects-parts-subparts-primitives hierarch…

Image GenerationImage Reconstructionmodel

Linguistically Driven Graph Capsule Network for Visual Question Reasoning

2020-03-23 · Qingxing Cao, Xiaodan Liang, Keze Wang, Liang Lin

Recently, studies of visual question answering have explored various architectures of end-to-end networks and achieved promising results on both natural and synthetic datasets, which require explicitly compositional reas…

Question AnsweringVisual Question AnsweringVisual Question Answering (VQA)

A Compositional and Interpretable Semantic Space

2015-05-01 · HLT 2015 5 · Brian Murphy, Leila Wehbe, Tom M. Mitchell, Partha P. Talukdar 외
Semantic Composition