EthicalML: Injecting Ethical and Legal Constraints into Machine Learning Models

EthicalML：将道德和法律约束注入机器学习模型

• 批准号: EP/P03442X/1
• 负责人: Novi Quadrianto
• 金额: $ 12.83万
• 依托单位: University of Sussex
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2017
• 资助国家: 英国
• 起止时间: 2017 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=EP%2FP03442X%2F1
• 关键词: EthicalML; Injecting; Ethical; Legal; Constraints

Our choice as to which movies to watch or novels to read can be influenced by suggestions made by machine learning (ML)-based recommender systems. However, there are some important scenarios where ML systems are deficient. Each of the following scenarios involves a situation where we wish to train an ML system so that it delivers a service. In each case, however, there is an important constraint that must be imposed on the operation of the ML system.Scenario 1: We want a system that will match submitted job applications to our list of academic vacancies. The system has to be non-discriminatory to minority groups. Scenario 2: We need an automated cancer diagnosis system based on biopsy images. We also have HIV test results, which can be used at training time but should not be collected from our new patients.Scenario 3: We wish to have a system that can aid us in deciding whether or not to approve a mortgage application. We need to understand the decision process and relate it to our checklist such as whether or not the applicant has an overdraft in the last three months and is on electoral roll. Scenario 1 asks an ML system to be fair in its decisions by being non-discriminatory with regards to, e.g., race, gender, and disability; scenario 2 requires an ML system to protect confidentiality of personal sensitive data; and scenario 3 demands transparency from an ML system by providing human-understandable decisions. Equipping ML models with ethical and legal constraints, scenarios 1-3, is a serious issue; without this, the future of ML is at risk. In the UK, this is recognized by the House of Commons Science and Technology Committee, which recommended an urgent formation of a Council of Data Ethics ("The Big Data Dilemma" report, 2016). Furthermore, since 2015, the Royal Society has started a policy project that looks at the social, legal, and ethical challenges associated with advancement in ML models and their use cases.Building ML models with fairness, confidentiality, and transparency constraints is an active research area, and disjoint frameworks are available for addressing each constraint. However, how to put them all together is not obvious. My long-term goal is to develop an ML framework with plug-and-play constraints that is able to handle any of the mentioned constraints, their combinations, and also new constraints that might be stipulated in the future.The proposed ML framework relies on instantiating ethical and legal constraints as privileged information. This privileged information is available at training time to better train a decision model and to make a decision model non-discriminatory, but it will not be accessible for future data at deployment time. For confidentiality constraints, personal confidential data such as HIV test results are the privileged information. For fairness constraints, protected characteristics such as race and gender are the privileged information. For transparency constraints, complex un-interpretable but highly discriminative features such as deep learning features are the privileged information.This project aims to develop an ML framework that produces accurate predictions and uncertainty estimates about its predictions while also complying with ethical and legal constraints. The key contributions of this proposal are: 1) a new privileged learning algorithm that overcomes limitations of existing methods by allowing to plug-and-play various constraints at deployment time, by being kernelized, by optimizing its hyperparameters, and by producing estimates of prediction uncertainty, 2) a scalable and automated inference that makes the new privileged learning algorithm easily applicable for any large scale learning problem such as binary classification, multi-class classification, and regression, and 3) an instantiation of the new algorithm for incorporating fairness, confidentiality, and transparency restrictions into ML models.

我们选择看哪部电影或读哪本小说，可能会受到基于机器学习（ML）的推荐系统的建议的影响。然而，在一些重要的场景中，ML系统存在缺陷。下面的每一个场景都涉及到一种情况，我们希望训练ML系统，以便它提供服务。然而，在每种情况下，都必须对ML系统的操作施加一个重要的约束。场景1：我们希望系统能够将提交的工作申请与我们的学术空缺列表相匹配。该制度必须不歧视少数群体。场景2：我们需要一个基于活检图像的自动癌症诊断系统。我们也有艾滋病毒检测结果，可以在培训时使用，但不应该从我们的新病人那里收集。场景3：我们希望有一个系统，可以帮助我们决定是否批准抵押贷款申请。我们需要了解决策过程，并将其与我们的清单相关联，例如申请人在过去三个月内是否透支，以及是否在选民名册上。场景1要求ML系统通过在以下方面不歧视来公平地做出决定，例如，场景2需要ML系统来保护个人敏感数据的机密性;场景3需要ML系统通过提供人类可理解的决策来实现透明度。为ML模型配备道德和法律的约束（场景1-3）是一个严重的问题;如果没有这一点，ML的未来将面临风险。在英国，这一点得到了下议院科学技术委员会的认可，该委员会建议紧急成立数据伦理理事会（“大数据困境”报告，2016年）。此外，自2015年以来，皇家学会已经开始了一个政策项目，该项目着眼于与ML模型及其用例的进步相关的社会，法律的和伦理挑战。构建具有公平性，机密性和透明度约束的ML模型是一个活跃的研究领域，并且可以使用不相交的框架来解决每个约束。然而，如何将它们放在一起并不明显。我的长期目标是开发一个具有即插即用约束的ML框架，该框架能够处理任何上述约束、其组合以及未来可能规定的新约束。拟议的ML框架依赖于实例化道德和法律的约束作为特权信息。这种特权信息在训练时可用，以更好地训练决策模型并使决策模型无歧视，但在部署时将无法访问未来的数据。出于保密性的限制，艾滋病毒检测结果等个人机密数据属于特许信息。对于公平约束，受保护的特征，如种族和性别是特权信息。对于透明度约束，复杂的不可解释但高度区分的特征（如深度学习特征）是特权信息。该项目旨在开发一个ML框架，该框架可以产生准确的预测和关于其预测的不确定性估计，同时也符合道德和法律的约束。该提案的主要贡献是：1）一种新的特权学习算法，其通过允许在部署时即插即用各种约束、通过内核化、通过优化其超参数以及通过产生预测不确定性的估计来克服现有方法的局限性，（二）一个可扩展的自动推理，使新的特权学习算法很容易适用于任何大规模的学习问题，如二进制分类，多类分类和回归，以及3）将公平性，机密性和透明度限制纳入ML模型的新算法的实例化。

项目成果

Discovering Fair Representations in the Data Domain

• DOI: 10.1109/cvpr.2019.00842
• 发表时间: 2018-10
• 期刊: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
• 作者: Novi Quadrianto;V. Sharmanska;Oliver Thomas

Causal Datasheet for Datasets: An Evaluation Guide for Real-World Data Analysis and Data Collection Design Using Bayesian Networks.

• DOI: 10.3389/frai.2021.612551
• 发表时间: 2021
• 期刊: Frontiers in artificial intelligence
• 影响因子: 4
• 作者: Butcher B;Huang VS;Robinson C;Reffin J;Sgaier SK;Charles G;Quadrianto N
• 通讯作者: Quadrianto N

Recycling Privileged Learning and Distribution Matching for Fairness

回收特权学习和分配匹配以实现公平

• 发表时间: 2017
• 期刊: Conference on Neural Information Processing Systems (NeurIPS -formerly NIPS)
• 作者: Quadrianto N

Low-variance Black-box Gradient Estimates for the Plackett-Luce Distribution

• DOI: 10.1609/aaai.v34i06.6572
• 发表时间: 2019-11
• 作者: Artyom Gadetsky;Kirill Struminsky;Christopher Robinson;Novi Quadrianto;D. Vetrov