FairLint-DL：面向深度学习软件公平性调试的IDE原生工具
FairLint-DL: An IDE-Native Tool for Fairness Debugging of Deep Learning Software
浏览论文内容
中文总结 AI 辅助
FairLint-DL是一个Visual Studio Code扩展，通过代理模型和QID指标实现训练前公平性检测，并在多个基准上高效发现偏差。
中文摘要 AI 辅助
现有的公平性分析工具主要作为训练后评估框架运行，要求从业者在评估偏差之前完成完整的模型开发生命周期。我们提出了FairLint-DL，一个Visual Studio Code扩展，通过实现公平性测试的左移方法，直接在表格数据集上进行训练前、IDE原生的偏差检测。FairLint-DL训练一个可配置的深度神经网络作为代理模型，并应用基于信息论的量化个体歧视（QID）指标。基于Shannon和最小熵，QID量化了受保护属性对预测的因果影响。该系统实现了一种两阶段梯度引导搜索算法来发现歧视性实例，一个因果调试流水线通过敏感性分析将偏差定位到特定网络层和神经元，以及使用SHAP和LIME的双重可解释性引擎进行特征级归因。在三个表格基准（Adult Census Income、German Credit和Bank Marketing）上的评估揭示了不同数据集间差异很大的公平性问题：在Adult上，96.0%的分析实例的QID高于0.1比特显著性阈值，平均QID为0.619比特，差异影响比为0.581，违反了五分之四法律规则。FairLint-DL在缓存模型上能在12秒内产生这些结果，证明了将公平性分析集成到开发者工作流程中而无需显著开销的可行性。
英文摘要
Existing fairness analysis tools predominantly operate as post-training evaluation frameworks, requiring practitioners to complete the full model development lifecycle before assessing bias. We present FairLint-DL, a Visual Studio Code extension that implements a shift-left approach to fairness testing by enabling pre-training, IDE-native bias detection directly on tabular datasets. FairLint-DL trains a configurable deep neural network as a proxy model and applies information-theoretic Quantitative Individual Discrimination (QID) metrics. Grounded in Shannon and min-entropy, QID quantifies the causal influence of protected attributes on predictions. The system implements a two-phase gradient-guided search algorithm for discovering discriminatory instances, a causal debugging pipeline that localizes bias to specific network layers and neurons via sensitivity analysis, and dual explainability engines using SHAP and LIME for feature-level attribution. Evaluation on three tabular benchmarks (Adult Census Income, German Credit, and Bank Marketing) reveals fairness concerns that vary widely across datasets: on Adult, 96.0% of analyzed instances exhibit QID above the 0.1-bit significance threshold, with a mean QID of 0.619 bits and a disparate impact ratio of 0.581, violating the four-fifths legal rule. FairLint-DL produces these results within 12 seconds on cached models, demonstrating the feasibility of integrating fairness analysis into the developer workflow without significant overhead.
发表机构
- University of Illinois at Chicago（伊利诺伊大学芝加哥分校）
机构由 AI 辅助整理，请以论文原文为准。
