Research Papers

In July 2023, New York City (NYC) initiated the first algorithm auditing system for commercial machine-learning systems. Local Law 144 (LL 144) mandates NYC-based employers using automated employment decision-making tools (AEDTs) in hiring to undergo annual bias audits conducted by an independent auditor. This paper examines lessons from LL 144 for other national algorithm auditing attempts. Through qualitative interviews with 16 experts and practitioners within the regime, we find that LL 144 has not effectively established an auditing regime. The law fails to clearly define key aspects, such as AEDTs and independent auditors, leading auditors, AEDT vendors, and companies using AEDTs to define the law's practical implementation in ways that failed to protect job applicants. Contributing factors include the law's flawed transparency-driven theory of change, industry lobbying narrowing the definition of AEDTs, practical and cultural challenges faced by auditors in accessing data, and wide disagreement over what constitutes a legitimate auditor, resulting in four distinct 'auditor roles.' We conclude with four recommendations for policymakers seeking to create similar bias auditing regimes, emphasizing clearer definitions, metrics, and increased accountability. By exploring LL 144 through the lens of auditors, our paper advances the evidence base around audit as an accountability mechanism, providing guidance for policymakers seeking to create similar regimes.

Established approaches to assuring safety-critical systems and software are difficult to apply to systems employing ML where there is no clear, pre-defined specification against which to assess validity. This problem is exacerbated by the "opaque" nature of ML where the learnt model is not amenable to human scrutiny. Explainable AI (XAI) methods have been proposed to tackle this issue by producing human-interpretable representations of ML models which can help users to gain confidence and build trust in the ML system. However, little work explicitly investigates the role of explainability for safety assurance in the context of ML development. This paper identifies ways in which XAI methods can contribute to safety assurance of ML-based systems. It then uses a concrete ML-based clinical decision support system, concerning weaning of patients from mechanical ventilation, to demonstrate how XAI methods can be employed to produce evidence to support safety assurance. The results are also represented in a safety argument to show where, and in what way, XAI methods can contribute to a safety case. Overall, we conclude that XAI methods have a valuable role in safety assurance of ML-based systems in healthcare but that they are not sufficient in themselves to assure safety.

This paper addresses the question of how to align AI systems with human values and situates it within a wider body of thought regarding technology and value. Far from existing in a vacuum, there has long been an interest in the ability of technology to 'lock-in' different value systems. There has also been considerable thought about how to align technologies with specific social values, including through participatory design-processes. In this paper we look more closely at the question of AI value alignment and suggest that the power and autonomy of AI systems gives rise to opportunities and challenges in the domain of value that have not been encountered before. Drawing important continuities between the work of the fairness, accountability, transparency and ethics community, and work being done by technical AI safety researchers, we suggest that more attention needs to be paid to the question of 'social value alignment' - that is, how to align AI systems with the plurality of values endorsed by groups of people, especially on the global level.

The Role of Cooperation in Responsible AI Development

Amanda Askell, Miles Brundage, Gillian Hadfield

In this paper, we argue that competitive pressures could incentivize AI companies to underinvest in ensuring their systems are safe, secure, and have a positive social impact. Ensuring that AI systems are developed responsibly may therefore require preventing and solving collective action problems between companies. We note that there are several key factors that improve the prospects for cooperation in collective action problems. We use this to identify strategies to improve the prospects for industry cooperation on the responsible development of AI.

Explaining Deep Classification of Time-Series Data with Learned Prototypes

Alan H. Gee, Diego Garcia-Olano, Joydeep Ghosh +1

The emergence of deep learning networks raises a need for explainable AI so that users and domain experts can be confident applying them to high-risk decisions. In this paper, we leverage data from the latent space induced by deep learning models to learn stereotypical representations or "prototypes" during training to elucidate the algorithmic decision-making process. We study how leveraging prototypes effect classification decisions of two dimensional time-series data in a few different settings: (1) electrocardiogram (ECG) waveforms to detect clinical bradycardia, a slowing of heart rate, in preterm infants, (2) respiration waveforms to detect apnea of prematurity, and (3) audio waveforms to classify spoken digits. We improve upon existing models by optimizing for increased prototype diversity and robustness, visualize how these prototypes in the latent space are used by the model to distinguish classes, and show that prototypes are capable of learning features on two dimensional time-series data to produce explainable insights during classification tasks. We show that the prototypes are capable of learning real-world features - bradycardia in ECG, apnea in respiration, and articulation in speech - as well as features within sub-classes. Our novel work leverages learned prototypical framework on two dimensional time-series data to produce explainable insights during classification tasks.

Large Language Models (LLMs) have transformed natural language processing (NLP) by enabling robust text generation and understanding. However, their deployment in sensitive domains like healthcare, finance, and legal services raises critical concerns about privacy and data security. This paper proposes a comprehensive framework for embedding trust mechanisms into LLMs to dynamically control the disclosure of sensitive information. The framework integrates three core components: User Trust Profiling, Information Sensitivity Detection, and Adaptive Output Control. By leveraging techniques such as Role-Based Access Control (RBAC), Attribute-Based Access Control (ABAC), Named Entity Recognition (NER), contextual analysis, and privacy-preserving methods like differential privacy, the system ensures that sensitive information is disclosed appropriately based on the user's trust level. By focusing on balancing data utility and privacy, the proposed solution offers a novel approach to securely deploying LLMs in high-risk environments. Future work will focus on testing this framework across various domains to evaluate its effectiveness in managing sensitive data while maintaining system efficiency.

Our research investigates the impact of Generative Artificial Intelligence (GAI) models, specifically text-to-image generators (T2Is), on the representation of non-Western cultures, with a focus on Indian contexts. Despite the transformative potential of T2Is in content creation, concerns have arisen regarding biases that may lead to misrepresentations and marginalizations. Through a community-centered approach and grounded theory analysis of 5 focus groups from diverse Indian subcultures, we explore how T2I outputs to English prompts depict Indian culture and its subcultures, uncovering novel representational harms such as exoticism and cultural misappropriation. These findings highlight the urgent need for inclusive and culturally sensitive T2I systems. We propose design guidelines informed by a sociotechnical perspective, aiming to address these issues and contribute to the development of more equitable and representative GAI technologies globally. Our work also underscores the necessity of adopting a community-centered approach to comprehend the sociotechnical dynamics of these models, complementing existing work in this space while identifying and addressing the potential negative repercussions and harms that may arise when these models are deployed on a global scale.

Gradient Boost Decision Trees (GBDT) is a powerful additive model based on tree ensembles. Its nature makes GBDT a black-box model even though there are multiple explainable artificial intelligence (XAI) models obtaining information by reinterpreting the model globally and locally. Each tree of the ensemble is a transparent model itself but the final outcome is the result of a sum of these trees and it is not easy to clarify. In this paper, a feature contribution method for GBDT is developed. The proposed method takes advantage of the GBDT architecture to calculate the contribution of each feature using the residue of each node. This algorithm allows to calculate the sequence of node decisions given a prediction. Theoretical proofs and multiple experiments have been carried out to demonstrate the performance of our method which is not only a local explicability model for the GBDT algorithm but also a unique option that reflects GBDTs internal behavior. The proposal is aligned to the contribution of characteristics having impact in some artificial intelligence problems such as ethical analysis of Artificial Intelligence (AI) and comply with the new European laws such as the General Data Protection Regulation (GDPR) about the right to explain and nondiscrimination.

Public attention towards explainability of artificial intelligence (AI) systems has been rising in recent years to offer methodologies for human oversight. This has translated into the proliferation of research outputs, such as from Explainable AI, to enhance transparency and control for system debugging and monitoring, and intelligibility of system process and output for user services. Yet, such outputs are difficult to adopt on a practical level due to a lack of a common regulatory baseline, and the contextual nature of explanations. Governmental policies are now attempting to tackle such exigence, however it remains unclear to what extent published communications, regulations, and standards adopt an informed perspective to support research, industry, and civil interests. In this study, we perform the first thematic and gap analysis of this plethora of policies and standards on explainability in the EU, US, and UK. Through a rigorous survey of policy documents, we first contribute an overview of governmental regulatory trajectories within AI explainability and its sociotechnical impacts. We find that policies are often informed by coarse notions and requirements for explanations. This might be due to the willingness to conciliate explanations foremost as a risk management tool for AI oversight, but also due to the lack of a consensus on what constitutes a valid algorithmic explanation, and how feasible the implementation and deployment of such explanations are across stakeholders of an organization. Informed by AI explainability research, we conduct a gap analysis of existing policies, leading us to formulate a set of recommendations on how to address explainability in regulations for AI systems, especially discussing the definition, feasibility, and usability of explanations, as well as allocating accountability to explanation providers.

Sample Selection for Fair and Robust Training

Yuji Roh, Kangwook Lee, Steven Euijong Whang +1

Fairness and robustness are critical elements of Trustworthy AI that need to be addressed together. Fairness is about learning an unbiased model while robustness is about learning from corrupted data, and it is known that addressing only one of them may have an adverse affect on the other. In this work, we propose a sample selection-based algorithm for fair and robust training. To this end, we formulate a combinatorial optimization problem for the unbiased selection of samples in the presence of data corruption. Observing that solving this optimization problem is strongly NP-hard, we propose a greedy algorithm that is efficient and effective in practice. Experiments show that our algorithm obtains fairness and robustness that are better than or comparable to the state-of-the-art technique, both on synthetic and benchmark real datasets. Moreover, unlike other fair and robust training baselines, our algorithm can be used by only modifying the sampling step in batch selection without changing the training algorithm or leveraging additional clean data.

In response to calls for greater interdisciplinary involvement from the social sciences and humanities in the development, governance, and study of artificial intelligence systems, this paper presents one sociologist's view on the problem of algorithmic bias and the reproduction of societal bias. Discussions of bias in AI cover much of the same conceptual terrain that sociologists studying inequality have long understood using more specific terms and theories. Concerns over reproducing societal bias should be informed by an understanding of the ways that inequality is continually reproduced in society -- processes that AI systems are either complicit in, or can be designed to disrupt and counter. The contrast presented here is between conservative and radical approaches to AI, with conservatism referring to dominant tendencies that reproduce and strengthen the status quo, while radical approaches work to disrupt systemic forms of inequality. The limitations of conservative approaches to class, gender, and racial bias are discussed as specific examples, along with the social structures and processes that biases in these areas are linked to. Societal issues can no longer be out of scope for AI and machine learning, given the impact of these systems on human lives. This requires engagement with a growing body of critical AI scholarship that goes beyond biased data to analyze structured ways of perpetuating inequality, opening up the possibility for radical alternatives.

Rule Extraction in Unsupervised Anomaly Detection for Model Explainability: Application to OneClass SVM

Alberto Barbado, Óscar Corcho, Richard Benjamins

OneClass SVM is a popular method for unsupervised anomaly detection. As many other methods, it suffers from the black box problem: it is difficult to justify, in an intuitive and simple manner, why the decision frontier is identifying data points as anomalous or non anomalous. Such type of problem is being widely addressed for supervised models. However, it is still an uncharted area for unsupervised learning. In this paper, we evaluate several rule extraction techniques over OneClass SVM models, as well as present alternative designs for some of those algorithms. Together with that, we propose algorithms to compute metrics related with eXplainable Artificial Intelligence (XAI) regarding the "comprehensibility", "representativeness", "stability" and "diversity" of the extracted rules. We evaluate our proposals with different datasets, including real-world data coming from industry. With this, our proposal contributes to extend XAI techniques to unsupervised machine learning models.