微小、始终在线且脆弱:设备上机器学习工作流中通过设计选择的偏差传播
Tiny, always-on and fragile: Bias propagation through design choices in on-device machine learning workflows
摘要:
数十亿个分布式、异构和资源受限的物联网设备部署了设备上的机器学习(ML),以实现对个人数据的私密、快速和离线推断。设备上ML高度依赖于上下文,对用户、使用、硬件和环境属性敏感。ML中的这种敏感性和偏倚倾向使得研究器械上设置的偏倚非常重要。我们的研究是对这一新兴领域偏见的首次调查之一,为构建更公平的设备上ML奠定了重要基础。我们应用一个软件工程镜头,通过在设备上ML工作流程中的设计选择来研究偏差的传播。首先,我们确定可靠性偏差的来源不公平,并提出了一种措施来量化它。然后,我们进行了实证实验的关键字发现任务,以显示复杂的和相互作用的技术设计选择放大和传播的可靠性偏差。我们的研究结果验证了在模型训练过程中所做的设计选择,如采样率和输入特征类型,以及优化模型的选择,如轻量级架构,修剪学习率和修剪稀疏度,可能会导致男性和女性群体的不同预测性能。根据我们的研究结果,我们建议工程师采取低投入策略来减轻设备上ML的偏见。
英文摘要:
Billions of distributed, heterogeneous and resource constrained IoT devices deploy on-device machine learning (ML) for private, fast and offline inference on personal data. On-device ML is highly context dependent, and sensitive to user, usage, hardware and environment attributes. This sensitivity and the propensity towards bias in ML makes it important to study bias in on-device settings. Our study is one of the first investigations of bias in this emerging domain, and lays important foundations for building fairer on-device ML. We apply a software engineering lens, investigating the propagation of bias through design choices in on-device ML workflows. We first identify reliability bias as a source of unfairness and propose a measure to quantify it. We then conduct empirical experiments for a keyword spotting task to show how complex and interacting technical design choices amplify and propagate reliability bias. Our results validate that design choices made during model training, like the sample rate and input feature type, and choices made to optimize models, like light-weight architectures, the pruning learning rate and pruning sparsity, can result in disparate predictive performance across male and female groups. Based on our findings we suggest low effort strategies for engineers to mitigate bias in on-device ML.
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。