arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2408.13438cs.CVcs.LG
收藏

通过视觉-语言偏好学习实现可解释的概念生成以理解神经网络的内部表示

Explainable Concept Generation through Vision-Language Preference Learning for Understanding Neural Networks' Internal Representations

Aditya Taparia, Som Sagar, Ransalu Senanayake

AI总结:

本文提出通过视觉-语言偏好学习生成概念图像集,解决传统方法需手动收集概念图像集的不足,提高神经网络内部表示理解的效率和可靠性。

摘要:

理解神经网络的内部表示有助于用户改进模型。基于概念的方法因能测试与特征属性无关的高层视觉“概念”而成为解释深度神经网络的流行选择。然而,概念解释方法需要从业者猜测并手动收集多个候选概念图像集,使过程繁琐且易遗漏重要概念。为解决此限制,本文将概念图像集创建视为图像生成问题。由于直接使用标准生成模型无法生成有意义的概念,我们设计了一种基于强化学习的偏好优化(RLPO)算法,通过微调视觉-语言生成模型以从近似文本描述中学习概念。通过一系列实验,我们展示了该方法在高效且可靠地表达多样化概念方面的能力,而这些概念通常难以手动构建。

英文摘要:

Understanding the inner representation of a neural network helps users improve models. Concept-based methods have become a popular choice for explaining deep neural networks post-hoc because, unlike most other explainable AI techniques, they can be used to test high-level visual "concepts" that are not directly related to feature attributes. For instance, the concept of "stripes" is important to classify an image as a zebra. Concept-based explanation methods, however, require practitioners to guess and manually collect multiple candidate concept image sets, making the process labor-intensive and prone to overlooking important concepts. Addressing this limitation, in this paper, we frame concept image set creation as an image generation problem. However, since naively using a standard generative model does not result in meaningful concepts, we devise a reinforcement learning-based preference optimization (RLPO) algorithm that fine-tunes a vision-language generative model from approximate textual descriptions of concepts. Through a series of experiments, we demonstrate our method's ability to efficiently and reliably articulate diverse concepts that are otherwise challenging to craft manually.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑