arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.13896cs.LGcs.AI
收藏

数据-算法-架构协同优化用于皮肤病变数据集上的公平神经网络

Data-Algorithm-Architecture Co-Optimization for Fair Neural Networks on Skin Lesion Dataset

George Mason University(乔治梅森大学);University of Pittsburgh Medical Center(匹兹堡大学医学中心);Guangdong Provincial People’s Hospital(广东省人民医院);University of Notre Dame(圣母大学);University of Pittsburgh(匹兹堡大学)

Yi Sheng, Junhuan Yang, Jinyang Li, James Alaina, Xiaowei Xu, Yiyu Shi, Jingtong Hu, Weiwen Jiang, Lei Yang

AI总结:

本文提出BiaslessNAS框架,通过数据-算法-架构协同优化,在皮肤病变数据集上实现公平神经网络,相比传统NAS,准确率提升2.55%,公平性提升65.50%。

摘要:

随着人工智能(AI)日益融入我们的日常生活,公平性已成为一个关键关注点,尤其是在医疗AI领域,其中数据集往往因社会因素(如边缘化群体代表性不足和社会经济障碍导致的数据收集困难)而反映出固有的偏见。缓解这些偏见的传统方法侧重于数据增强和开发公平感知的训练算法。然而,本文认为神经网络的架构作为机器学习(ML)的核心组成部分,在确保公平性方面起着至关重要的作用。我们证明,有效解决公平性问题需要一种整体方法,同时考虑数据、算法和架构。利用自动化机器学习(AutoML)技术,特别是神经架构搜索(NAS),我们引入了一个新颖的框架BiaslessNAS,旨在分析皮肤病变数据集时实现公平的结果。BiaslessNAS在NAS过程的每个阶段都纳入公平性考虑,从而识别出不仅更准确而且显著更公平的神经网络。我们的实验表明,与传统NAS方法相比,BiaslessNAS在准确率上提高了2.55%,在公平性上改善了65.50%,这强调了将公平性融入神经网络架构对于在医疗AI应用中取得更好结果的重要性。

英文摘要:

As Artificial Intelligence (AI) increasingly integrates into our daily lives, fairness has emerged as a critical concern, particularly in medical AI, where datasets often reflect inherent biases due to social factors like the underrepresentation of marginalized communities and socioeconomic barriers to data collection. Traditional approaches to mitigating these biases have focused on data augmentation and the development of fairness-aware training algorithms. However, this paper argues that the architecture of neural networks, a core component of Machine Learning (ML), plays a crucial role in ensuring fairness. We demonstrate that addressing fairness effectively requires a holistic approach that simultaneously considers data, algorithms, and architecture. Utilizing Automated ML (AutoML) technology, specifically Neural Architecture Search (NAS), we introduce a novel framework, BiaslessNAS, designed to achieve fair outcomes in analyzing skin lesion datasets. BiaslessNAS incorporates fairness considerations at every stage of the NAS process, leading to the identification of neural networks that are not only more accurate but also significantly fairer. Our experiments show that BiaslessNAS achieves a 2.55% increase in accuracy and a 65.50% improvement in fairness compared to traditional NAS methods, underscoring the importance of integrating fairness into neural network architecture for better outcomes in medical AI applications.

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。

↑