We propose a learning setting in which unlabeled data is free, and the cost of a label depends on its value, which is not known in advance. We study binary classification in an extreme case, where the algorithm only pays for negative labels. Our motivation are applications such as fraud detection, in which investigating an honest transaction should be avoided if possible. We term the setting auditing, and consider the auditing complexity of an algorithm: the number of negative labels the algorithm requires in order to learn a hypothesis with low relative error. We design auditing algorithms for simple hypothesis classes (thresholds and rectangles), and show that with these algorithms, the auditing complexity can be significantly lower than the active label complexity. We also discuss a general competitive approach for auditing and possible modifications to the framework.
Auditing Near-Optimal Policies Can Be Exponentially Hard: Conditional Query Lower Bounds via Occupancy Rashomon Capacity
Department of Computer Science, Iowa State University（计算机科学系，爱荷华州立大学）；Department of Civil, Construction & Environmental Engineering, Iowa State University（土木、建设与环境工程系，爱荷华州立大学）
Carnegie Mellon University（卡内基梅隆大学）；University of Arizona（亚利桑那大学）
The fast spreading adoption of machine learning (ML) by companies across industries poses significant regulatory challenges. One such challenge is scalability: how can regulatory bodies efficiently audit these ML models, ensuring that they are fair? In this paper, we initiate the study of query-based auditing algorithms that can estimate the demographic parity of ML models in a query-efficient manner. We propose an optimal deterministic algorithm, as well as a practical randomized, oracle-efficient algorithm with comparable guarantees. Furthermore, we make inroads into understanding the optima…
Université Libre de Bruxelles（布鲁塞尔自由大学）；Athena Research Center（雅典娜研究中心）；Tampere University（坦佩雷大学）
This paper studies algorithmic fairness when the protected attribute is location. To handle protected attributes that are continuous, such as age or income, the standard approach is to discretize the domain into predefined groups, and compare algorithmic outcomes across groups. However, applying this idea to location raises concerns of gerrymandering and may introduce statistical bias. Prior work addresses these concerns but only for regularly spaced locations, while raising other issues, most notably its inability to discern regions that are likely to exhibit spatial unfairness. Similar to es…
Missouri University of Science and Technology（密苏里科技大学）
Bias evaluation in machine-learning based services (MLS) based on traditional algorithmic fairness notions that rely on comparative principles is practically difficult, making it necessary to rely on human auditor feedback. However, in spite of taking rigorous training on various comparative fairness notions, human auditors are known to disagree on various aspects of fairness notions in practice, making it difficult to collect reliable feedback. This paper offers a paradigm shift to the domain of algorithmic fairness via proposing a new fairness notion based on the principle of non-comparative…
From Sampled Outcomes to Capability Distributions: Rethinking Supervision for LLM Routing
School of Artificial Intelligence, Nanjing University（南京大学人工智能学院）；National Key Laboratory for Novel Software Technology, Nanjing University（南京大学新型软件技术国家重点实验室）；Hong Kong University of Science and Technology（香港科学与技术大学）；Frontier Robotics（前沿机器人）
ICON decomposition: Multivariate concept auditing reveals shortcuts that neural networks rely on
Hertie Institute for AI in Brain Health, University of Tübingen（蒂宾根大学赫蒂脑健康人工智能研究所）；Charité - Universitätsmedizin Berlin（柏林夏里特医学院）；Humboldt-Universität zu Berlin（柏林洪堡大学）；Tübingen AI Center, University of Tübingen（蒂宾根大学蒂宾根人工智能中心）；German Center for Mental Health (DZPG)（德国心理健康中心）；Fraunhofer Heinrich Hertz Institute（弗劳恩霍夫海因里希赫兹研究所）；Technische Universität Berlin（柏林工业大学）
Haverford College（哈弗福德学院）；University of Arizona（亚利桑那大学）；University of Utah（犹他大学）
Data-trained predictive models see widespread use, but for the most part they are used as black boxes which output a prediction or score. It is therefore hard to acquire a deeper understanding of model behavior, and in particular how different features influence the model prediction. This is important when interpreting the behavior of complex models, or asserting that certain problematic attributes (like race or gender) are not unduly influencing decisions. In this paper, we present a technique for auditing black-box models, which lets us study the extent to which existing models take advantag…
Targeted Active Learning for Preference-Based Treatment Effects on Multivariate Outcomes
Theremia；Paris Brain Institute (ICM)（巴黎脑科学研究所（ICM））；Sorbonne Université（索邦大学）；Inserm（法国国家健康与医学研究院）；CNRS（法国国家科学研究中心）；AP-HP（巴黎公共援助医院）；Inria（法国国家信息与自动化研究所）；Université de Montpellier（蒙彼利埃大学）；Université Paris Cité（巴黎西岱大学）；Centre for Research in Epidemiology and Statistics (CRESS)（流行病学与统计学研究中心（CRESS））
Consiglio Nazionale delle Ricerche（意大利国家研究委员会）；Istituto di Sistemi e Tecnologie Industriali per il Manifatturiero Avanzato（先进制造工业系统与技术研究所）；Scuola Universitaria Professionale della Svizzera Italiana（瑞士意大利语区应用科学大学）；Università della Svizzera Italiana（瑞士意大利语区大学）；Istituto Dalle Molle di studi sull’Intelligenza Artificiale（达勒莫勒人工智能研究所）；University of Bologna（博洛尼亚大学）
Black-box optimization refers to the optimization problem whose objective function and/or constraint sets are either unknown, inaccessible, or non-existent. In many applications, especially with the involvement of humans, the only way to access the optimization problem is through performing physical experiments with the available outcomes being the preference of one candidate with respect to one or many others. Accordingly, the algorithm so-called Active Preference Learning has been developed to exploit this specific information in constructing a surrogate function based on standard radial bas…
Ontology-Amplified Distillation and Contextuality Auditing for Sovereign Enterprise Language Models: A Combined Proof-of-Mechanism and Negative-Results Method Study
