arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.12443cs.AIcs.LG
收藏

一种可问责和可复制的联邦学习:一种FactSheets方法

Towards an Accountable and Reproducible Federated Learning: A FactSheets Approach

IBM Research(IBM研究院);fortiss(fortiss研究所)

Nathalie Baracaldo, Ali Anwar, Mark Purcell, Ambrish Rawat, Mathieu Sinn, Bashar Altakrouri, Dian Balta, Mahdi Sellami, Peter Kuhn, Ulrich Schopp, Matthias Buchinger

摘要:

联邦学习(FL)是一种基于分散和私有数据的模型共享训练的新范式。在道德准则方面,FL在隐私方面很有希望,但需要在透明度和可信度方面表现出色。特别是,FL必须解决有关各方的责任问题及其遵守规则、法律和原则的问题。我们引入AF^2框架,在这里我们通过将可验证的主张与明显篡改的事实融合在一起,将FL与问责制结合起来,形成可重复的论点。我们以AI FactSheets为基础,将透明度和可信度融入AI生命周期,并将其扩展到FL中的动态和嵌套事实以及复杂的模型组合。基于我们的方法,审核员可以验证,复制和认证FL流程。这可以直接应用于实践中,以应对人工智能工程和道德的挑战。

英文摘要:

Federated Learning (FL) is a novel paradigm for the shared training of models based on decentralized and private data. With respect to ethical guidelines, FL is promising regarding privacy, but needs to excel vis-à-vis transparency and trustworthiness. In particular, FL has to address the accountability of the parties involved and their adherence to rules, law and principles. We introduce AF^2 Framework, where we instrument FL with accountability by fusing verifiable claims with tamper-evident facts, into reproducible arguments. We build on AI FactSheets for instilling transparency and trustworthiness into the AI lifecycle and expand it to incorporate dynamic and nested facts, as well as complex model compositions in FL. Based on our approach, an auditor can validate, reproduce and certify a FL process. This can be directly applied in practice to address the challenges of AI engineering and ethics.

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。

↑