提出可解释的无监督图像分类模型，提升泛化与人类可理解性
Towards Better Generalization and Interpretability in Unsupervised Concept-Based Models
- 将概念建模为伯努利潜空间中的随机变量，减少概念数量同时保持性能
- 在泛化能力上超越现有无监督模型，接近黑箱模型表现
- 概念更贴近人类理解，且通过局部线性组合保持可解释性
为提升深度神经网络的可信度，关键在于增强对其决策过程的理解。本文提出一种新型无监督概念基础模型——可学习概念基础模型（LCBM），将概念建模为伯努利潜空间中的随机变量。与需要大量人工标注或可扩展性受限的传统方法不同，该方法使用较少概念数量，同时不损失性能。实验表明，LCBM在泛化能力上优于现有无监督概念模型，且几乎达到黑箱模型的性能水平。所提出的概念表示增强了信息保留，并更符合人类理解。用户研究进一步证明，发现的概念对人类更具直观可解释性。此外，尽管使用概念嵌入，仍通过局部线性组合保持模型可解释性。
原文摘要 · Abstract (English)
To increase the trustworthiness of deep neural networks, it is critical to improve the understanding of how they make decisions. This paper introduces a novel unsupervised concept-based model for image classification, named Learnable Concept-Based Model (LCBM) which models concepts as random variables within a Bernoulli latent space. Unlike traditional methods that either require extensive human supervision or suffer from limited scalability, our approach employs a reduced number of concepts without sacrificing performance. We demonstrate that LCBM surpasses existing unsupervised concept-based models in generalization capability and nearly matches the performance of black-box models. The proposed concept representation enhances information retention and aligns more closely with human understanding. A user study demonstrates the discovered concepts are also more intuitive for humans to interpret. Finally, despite the use of concept embeddings, we maintain model interpretability by means of a local linear combination of concepts.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。