Machine learning researchers distinguish between reinforcement learning and supervised learning and refer to reinforcement learning systems as “agents”. This paper vindicates the claim that systems trained by reinforcement learning are agents while those trained by supervised learning are not. Systems of both kinds satisfy Dretske’s criteria for agency, because they both learn to produce outputs selectively in response to inputs. However, reinforcement learning is sensitive to the instrumental value of outputs, giving rise (...) to systems which exploit the effects of outputs on subsequent inputs to achieve good performance over episodes of interaction with their environments. Supervised learning systems, in contrast, merely learn to produce better outputs in response to individual inputs. (shrink)

Machine learning (ML) shows strong performance in making accurate inferences from massive, high-dimensional data. Many scientists turn to this new tool when traditional inferential procedures cannot deal with overly messy data and complex target phenomena. One example is galactic archaeology, a branch of astronomy that aims to unravel the epic history of the Milky Way using the present snapshot of stars with only a handful of physical parameters. Historical inference in galactic archaeology is difficult due to the uncertainty (...) of what counts as evidence of past events. While ML is expected to help address such challenging situations, it comes with its own epistemological limitations. Two issues, theory-ladenness and non-uniqueness, can undermine the evidential power of ML outcomes in scientific inferences. ML’s success in aiding Galactic archaeology teaches lessons about how it can serve key epistemic functions for the historical sciences, unaffected by those two issues. I show that ML can empower the evidence-gathering procedure in science, not by outputting data to be directly taken as evidence, but as evidential constraints: ML partitions and structures data for it to be subsequently interpreted as evidence with the help of domain knowledge and inferential procedures. Used in this way, ML not only provides a solution to the ambiguity of historical evidence but also avoids its own epistemic limitations. This use of ML can be generalized to other historical sciences that engage with big data and automated inference. (shrink)

In the past few years, machine learning (ML) tools have been implemented with success in the medical context. However, several practitioners have raised concerns about the lack of transparency—at the algorithmic level—of many of these tools; and solutions from the field of explainable AI (XAI) have been seen as a way to open the ‘black box’ and make the tools more trustworthy. Recently, Alex London has argued that in the medical context we do not need machine (...) class='Hi'>learning tools to be interpretable at the algorithmic level to make them trustworthy, as long as they meet some strict empirical desiderata. In this paper, we analyse and develop London’s position. In particular, we make two claims. First, we claim that London’s solution to the problem of trust can potentially address another problem, which is how to evaluate the reliability of ML tools in medicine for regulatory purposes. Second, we claim that to deal with this problem, we need to develop London’s views by shifting the focus from the opacity of algorithmic details to the opacity of the way in which ML tools are trained and built. We claim that to regulate AI tools and evaluate their reliability, agencies need an explanation of how ML tools have been built, which requires documenting and justifying the technical choices that practitioners have made in designing such tools. This is because different algorithmic designs may lead to different outcomes, and to the realization of different purposes. However, given that technical choices underlying algorithmic design are shaped by value-laden considerations, opening the black box of the design process means also making transparent and motivating (technical and ethical) values and preferences behind such choices. Using tools from philosophy of technology and philosophy of science, we elaborate a framework showing how an explanation of the training processes of ML tools in medicine should look like. (shrink)

When courts started publishing judgements, big data analysis within the legal domain became possible. By taking data from the European Court of Human Rights as an example, we investigate how natural language processing tools can be used to analyse texts of the court proceedings in order to automatically predict judicial decisions. With an average accuracy of 75% in predicting the violation of 9 articles of the European Convention on Human Rights our approach highlights the potential of machine learning (...) approaches in the legal domain. We show, however, that predicting decisions for future cases based on the cases from the past negatively impacts performance. Furthermore, we demonstrate that we can achieve a relatively high classification performance when predicting outcomes based only on the surnames of the judges that try the case. (shrink)

Prediction-based decisions, which are often made by utilizing the tools of machine learning, influence nearly all facets of modern life. Ethical concerns about this widespread practice have given rise to the field of fair machine learning and a number of fairness measures, mathematically precise definitions of fairness that purport to determine whether a given prediction-based decision system is fair. Following Reuben Binns (2017), we take ‘fairness’ in this context to be a placeholder for a variety of (...) normative egalitarian considerations. We explore a few fairness measures to suss out their egalitarian roots and evaluate them, both as formalizations of egalitarian ideas and as assertions of what fairness demands of predictive systems. We pay special attention to a recent and popular fairness measure, counterfactual fairness, which holds that a prediction about an individual is fair if it is the same in the actual world and any counterfactual world where the individual belongs to a different demographic group (cf. Kusner et al. (2018)). (shrink)

Current methods of operationalizing concepts of misinformation in machine learning are often problematic given idiosyncrasies in their success conditions compared to other models employed in the natural and social sciences. The intrinsic value-ladenness of misinformation and the dynamic relationship between citizens' and social scientists' concepts of misinformation jointly suggest that both the construct legitimacy and the construct validity of these models needs to be assessed via more democratic criteria than has previously been recognized.

Parameterization and parameter tuning are central aspects of climate modeling, and there is widespread consensus that these procedures involve certain subjective elements. Even if the use of these subjective elements is not necessarily epistemically problematic, there is an intuitive appeal for replacing them with more objective (automated) methods, such as machine learning. Relying on several case studies, we argue that, while machine learning techniques may help to improve climate model parameterization in several ways, they still require (...) expert judgment that involves subjective elements not so different from the ones arising in standard parameterization and tuning. The use of machine learning in parameterizations is an art as well as a science and requires careful supervision. (shrink)

Machine learning algorithms may radically improve our ability to diagnose and treat disease. For moral, legal, and scientific reasons, it is essential that doctors and patients be able to understand and explain the predictions of these models. Scalable, customisable, and ethical solutions can be achieved by working together with relevant stakeholders, including patients, data scientists, and policy makers.

Machine learning is the study of algorithms that enable computers to improve their performance and increase their knowledge base. Research in machine learning has taken place since the beginning of artificial intelligence in the mid‐1950s. The first notable success was Arthur Samuel's program that learned to play checkers well enough to beat skilled humans. The program estimated the best move in a situation by using a mathematical function whose sixteen parameters describe board positions, and it improved (...) its performance by adjusting weights on the parameters in order to generalize about what moves were successful. (shrink)

Recent controversies about the level of replicability of behavioral research analyzed using statistical inference have cast interest in developing more efficient techniques for analyzing the results of psychological experiments. Here we claim that complementing the analytical workflow of psychological experiments with Machine Learning-based analysis will both maximize accuracy and minimize replicability issues. As compared to statistical inference, ML analysis of experimental data is model agnostic and primarily focused on prediction rather than inference. We also highlight some potential pitfalls (...) resulting from adoption of Machine Learning based experiment analysis. If not properly used it can lead to over-optimistic accuracy estimates similarly observed using statistical inference. Remedies to such pitfalls are also presented such and building model based on cross validation and the use of ensemble models. ML models are typically regarded as black boxes and we will discuss strategies aimed at rendering more transparent the predictions. (shrink)

Just as humans can draw conclusions responsibly or irresponsibly, so too can computers. Machine learning systems that have been trained on data sets that include irresponsible judgments are likely to yield irresponsible predictions as outputs. In this paper I focus on a particular kind of inference a computer system might make: identification of the intentions with which a person acted on the basis of photographic evidence. Such inferences are liable to be morally objectionable, because of a way in (...) which they are presumptuous. After elaborating this moral concern, I explore the possibility that carefully procuring the training data for image recognition systems could ensure that the systems avoid the problem. The lesson of this paper extends beyond just the particular case of image recognition systems and the challenge of responsibly identifying a person’s intentions. Reflection on this particular case demonstrates the importance (as well as the difficulty) of evaluating machine learning systems and their training data from the standpoint of moral considerations that are not encompassed by ordinary assessments of predictive accuracy. (shrink)

Simple idealized models seem to provide more understanding than opaque, complex, and hyper-realistic models. However, an increasing number of scientists are going in the opposite direction by utilizing opaque machine learning models to make predictions and draw inferences, suggesting that scientists are opting for models that have less potential for understanding. Are scientists trading understanding for some other epistemic or pragmatic good when they choose a machine learning model? Or are the assumptions behind why minimal models (...) provide understanding misguided? In this paper, using the case of deep neural networks, I argue that it is not the complexity or black box nature of a model that limits how much understanding the model provides. Instead, it is a lack of scientific and empirical evidence supporting the link that connects a model to the target phenomenon that primarily prohibits understanding. (shrink)

Decisions based on algorithmic, machine learning models can be unfair, reproducing biases in historical data used to train them. While computational techniques are emerging to address aspects of these concerns through communities such as discrimination-aware data mining and fairness, accountability and transparency machine learning, their practical implementation faces real-world challenges. For legal, institutional or commercial reasons, organisations might not hold the data on sensitive attributes such as gender, ethnicity, sexuality or disability needed to diagnose and mitigate (...) emergent indirect discrimination-by-proxy, such as redlining. Such organisations might also lack the knowledge and capacity to identify and manage fairness issues that are emergent properties of complex sociotechnical systems. This paper presents and discusses three potential approaches to deal with such knowledge and information deficits in the context of fairer machine learning. Trusted third parties could selectively store data necessary for performing discrimination discovery and incorporating fairness constraints into model-building in a privacy-preserving manner. Collaborative online platforms would allow diverse organisations to record, share and access contextual and experiential knowledge to promote fairness in machine learning systems. Finally, unsupervised learning and pedagogically interpretable algorithms might allow fairness hypotheses to be built for further selective testing and exploration. Real-world fairness challenges in machine learning are not abstract, constrained optimisation problems, but are institutionally and contextually grounded. Computational fairness tools are useful, but must be researched and developed in and with the messy contexts that will shape their deployment, rather than just for imagined situations. Not doing so risks real, near-term algorithmic harm. (shrink)

Informed consent is at the core of the clinical relationship. With the introduction of machine learning (ML) in healthcare, the role of informed consent is challenged. This paper addresses the issue of whether patients must be informed about medical ML applications and asked for consent. It aims to expose the discrepancy between ethical and practical considerations, while arguing that this polarization is a false dichotomy: in reality, ethics is applied to specific contexts and situations. Bridging this gap and (...) considering the whole picture is essential for advancing the debate. In the light of the possible future developments of the situation and the technologies, as well as the benefits that informed consent for ML can bring to shared decision-making, the present analysis concludes that it is necessary to prepare the ground for a possible future requirement of informed consent for medical ML. (shrink)

What is the relationship between ideas of sameness and difference for machine learning and AI? Algorithms are often understood to participate in the continual displacement of the different and heterogeneous in society in favour of sameness, of that which is socio-politically similar and proximate. In contrast to this prevalent emphasis on sameness, however, this paper argues that there is a nascent heterophilic logic underpinning the intersection of synthetic data and machine learning, a move towards actively generating (...) differences and heterogeneous data attributes to train, fine-tune, and optimize algorithms. Yet, these synthetic attribute data are nonetheless always machine compatible, devoid of their socio-cultural dynamics and tensions. As such, through a critical examination of three core dimensions of this emergent politics of difference of synthetic data – disentanglement, compositionality, and normativity – the paper argues that this has the potential to ultimately undercut a politics of intervention that seeks to foreground the systemic unfairness and violence of machine learning. (shrink)

I consider three aspects in which machine learning and philosophy of science can illuminate each other: methodology, inductive simplicity and theoretical terms. I examine the relations between the two subjects and conclude by claiming these relations to be very close.

This article examines what the rise in machine learning (ML) systems might mean for social theory. Focusing on financial markets, in which algorithmic securities trading founded on ML-based decision-making is gaining traction, I discuss the extent to which established sociological notions remain relevant or demand a reconsideration when applied to an ML context. I argue that ML systems have some capacity for agency and for engaging in forms of collective machine behaviour, in which ML systems interact with (...) other machines. However, ML-based collective machine behaviour is irreducible to human decision-making and thereby challenges established sociological notions of financial markets (including that of embeddedness). I argue that such behaviour can nonetheless be analysed through an adaptation of sociological theories of interaction and collective behaviour. (shrink)

This study focuses on a traditional business ethics question and aims to use advanced techniques to improve the performance of corporate fraud prediction. Based on the GONE framework, we adopt the machine learning model to predict the occurrence of corporate fraud in China. We first identify a comprehensive set of fraud-related variables and organize them into each category (i.e., Greed, Opportunity, Need, and Exposure) of the GONE framework. Among the six machine learning models tested, the Random (...) Forest (RF) model outperforms the other five models in corporate fraud prediction. Based on the RF model, we show that Exposure variables play a more important role in predicting corporate fraud than other input variables. These results highlight the importance of Exposure variables in corporate fraud prediction and promote the practical use of the machine learning model in solving business ethics questions. (shrink)

Issues of reliability are claiming center-stage in the epistemology of machine learning. This paper unifies different branches in the literature and points to promising research directions, whilst also providing an accessible introduction to key concepts in statistics and machine learning – as far as they are concerned with reliability.

This paper addresses two methodological paths in Artificial Intelligence: the paths of Babbage and anti-Cartesianism. While those researchers who have followed the latter have attempted to reverse the Cartesian dictum according to which machines cannot think in principle, Babbage’s path, which has been partially neglected, implies that the replacement of humans—and not the creation of minds—should provide the foundation of AI. In view of the examined paths, the claim that we support here is this: in line with Babbage, AI researchers (...) have recently concentrated upon the replacement of human labour, and thus upon the creation of Machine Learning systems. After presenting and analysing the paths, we characterise Machine Learning via its developments and an illustrative example. Then, we put forward an argument that shows that total replacement of human labour will not be feasible for practical and conceptual reasons despite the successful developments in recent AI systems. Our discussion finally leads to optimism and awareness: AI’s advances allow humans to dedicate themselves to higher level tasks, but these advances also require that we be vigilant about the responsibilities granted to ML-based systems. (shrink)

As we write this research paper, we notice an explosion in popularity of machine learning in numerous fields (ranging from governance, education, and management to criminal justice, fraud detection, and internet of things). In this contribution, rather than focusing on any of those fields, which have been well-reviewed already, we decided to concentrate on a series of more recent applications of deep learning models and technologies that have only recently gained significant track in the relevant literature. These (...) applications are concerned with artistic production (Sect. 2.1), the writing process (Sect. 2.2), music production (Sect. 2.3), text recognition and attribution (Sect. 2.4). After reviewing and analyzing the positive contributions as well as some of the major limitations of these technologies in each of those fields, we critically reflect (Sect. 3) on how their widespread implementation may affect humans and their creativity. In Sect. 4, we notice that deep learning models are here to stay; so, rather than embracing a negative or pessimistic stance with respect to their future applications in creative domains, we suggest a balanced approach for their assessment and beneficial usage. Finally (Sect. 5), we conclude by summarising what we have achieved and by pointing out possible future research directions. (shrink)

This article examines how machine learning and natural language processing impact human emotional development. The main thesis is that a reciprocal relationship exists by programming machines with specific causal categories, we shape how they interact with us, and in turn, these machines influence human environments and emotional experiences. This mutual shaping can encourage reductionist views by both humans and machines. To address ethical concerns arising from this interaction, we emphasize the need to recognize the context and relativity of (...) emotion recognition technology. Emotions are inherently indeterminate and cannot be fully captured by any single framework. (shrink)

Machine learning and essentialism have been connected in the past by various researchers, in order to state that the main paradigm in machine learning processes is equivalent to choosing the “essential” attributes for the machine to search for. Our goal in this paper is to show that there are connections between machine learning and essentialism, but only for some kinds of machine learning, and often not including deep learning methods. Similarity-based (...) approaches, more connected to the overall prototype theory, spanning from psychology and linguistics, seem more suited for pattern recognition and complex deep-learning issues, while for classification problems, mostly for unsupervised learning, essentialism seems like the best choice. In order to illustrate the difference better, we will connect both paths to their sources in other disciplines and see how human psychology influences our decision in machine-learning modeling as well. This leads to a philosophically very interesting consequence: even in the setting of supervised machine learning, essences are not present in data, but in targets, which in turn means that the categories which purport to be essences are in fact human-made, and hand-coded in the targets. The success of machine learning, therefore, does not give any substantial evidence for the independent existence of essential properties. Our stance here is to state that neither the existence nor the lack of “essential” properties in machine learning can lead to metaphysical, i.e., ontological claims. (shrink)

Machine learning (ML) models recently led to major breakthroughs in predictive tasks in the natural sciences. Yet their benefits for the social sciences are less evident, as even high-profile studies on the prediction of life trajectories have shown to be largely unsuccessful – at least when measured in traditional criteria of scientific success. This paper tries to shed light on this remarkable performance gap. Comparing two social science case studies to a paradigm example from the natural sciences, we (...) argue that, in addition to explanation, prediction is an important goal of social science – and we identify constraints that impede pure ML prediction from being successful in that field. As a remedy, we outline elements of an integrative modelling approach that combines explanatory models and predictive ML models. (shrink)

Large language models (LLMs) such as OpenAI’s ChatGPT reflect, and can potentially perpetuate, social biases in language use. Conceptual engineering aims to revise our concepts to eliminate such bias. We show how machine learning and conceptual engineering can be fruitfully brought together to offer new insights to both conceptual engineers and LLM designers. Specifically, we suggest that LLMs can be used to detect and expose bias in the prototypes associated with concepts, and that LLM de-biasing can serve conceptual (...) engineering projects that aim to revise such conceptual prototypes. At present, these de-biasing techniques primarily involve approaches requiring bespoke interventions based on choices of the algorithm’s designers. Thus, conceptual engineering through de-biasing will include making choices about what kind of normative training an LLM should receive, especially with respect to different notions of bias. This offers a new perspective on what conceptual engineering involves and how it can be implemented. And our conceptual engineering approach also offers insight, to those engaged in LLM de-biasing, into the normative distinctions that are needed for that work. (shrink)

There has been an increased focus within the AI ethics literature on questions of power, reflected in the ideal of accountability supported by many Responsible AI guidelines. While this recent debate points towards the power asymmetry between those who shape AI systems and those affected by them, the literature lacks normative grounding and misses conceptual clarity on how these power dynamics take shape. In this paper, I develop a workable conceptualization of said power dynamics according to Cristiano Castelfranchi’s conceptual framework (...) of power and argue that end-users depend on a system’s developers and users, because end-users rely on these systems to satisfy their goals, constituting a power asymmetry between developers, users and end-users. I ground my analysis in the neo-republican moral wrong of domination, drawing attention to legitimacy concerns of the power-dependence relation following from the current lack of accountability mechanisms. I illustrate my claims on the basis of a risk-prediction machine learning system, and propose institutional and project-specific solutions to mitigate domination. (shrink)

Neurodegenerative Parkinson's Disease (PD) is one of the common incurable diseases among the elderly. Clinical assessments are characterized as standardized means for PD diagnosis. However, relying on medical evaluation of a patient's status can be subjective to physicians' experience, making the assessment process susceptible to human errors. The use of ICT-based tools for capturing the status of patients with PD can provide more objective and quantitative metrics. In this vein, the Personalized Serious Game Suite (PGS) and intelligent Motor Assessment Tests (...) (iMAT), produced within the i-PROGNOSIS European project (www.i-prognosis.eu), are explored in the current study. More specifically, data from 27 patients with PD at Stage 1 (9) and Stage 3 (18) produced from their interaction with PGS/iMAT are analyzed. Five feature vector (FV) scenarios are set, including features from PGS or iMAT scores or their combination, after also taking into consideration the age of patients with PD. These FVs are fed into three machine learning classifiers, i.e., K-Nearest Neighbor (KNN), Support Vector Machines (SVM), and Random Forest (RF), to infer the stage of each patient with PD. A Leave-One-Out Cross-Validation (LOOCV) method is adopted for testing the classification performance. The experimental results show that a high (>90%) classification accuracy is achieved from both data sources (PGS/iMAT), justifying the effectiveness of PGS/iMAT to efficiently reflect the motor skill status of patients with PD and further potentiating PGS/iMAT enhancement with a machine learning a part to infer for the stage of patients with PD. Clearly, this integrated approach provides new opportunities for remote monitoring of the stage of patients with PD, contributing to a more efficient organization and set up of personalized interventions. (shrink)

Many foundational concepts in machine learning have been criticized as inadequate. Philosophers have therefore taken it upon themselves to sort out the conceptual terrain—with conceptual engineering being the method of choice. This paper takes a step back to provide theoretical and methodological grounding for future work on conceptual engineering in machine learning. To this end, we consider the functional roles of concepts in machine learning, the underlying causes and types of deficiency, and map out (...) criteria for the successful propagation of reengineered concepts within and beyond the machine learning community. Moreover, we discuss how the space of viable conceptual revisions in machine learning is constrained by the need for operationalization, and how tensions can occur between the sociopolitical desirability and the computational implementability of relevant conceptual engineering projects. Overall, our goal is to delineate how conceptual work in philosophy ought to be if the goal is for our contribution to permeate through the science and practice of machine learning. (shrink)

We argue why interpretability should have primacy alongside empiricism for several reasons: first, if machine learning models are beginning to render some of the high-risk healthcare decisions instead of clinicians, these models pose a novel medicolegal and ethical frontier that is incompletely addressed by current methods of appraising medical interventions like pharmacological therapies; second, a number of judicial precedents underpinning medical liability and negligence are compromised when ‘autonomous’ ML recommendations are considered to be en par with human instruction (...) in specific contexts; third, explainable algorithms may be more amenable to the ascertainment and minimisation of biases, with repercussions for racial equity as well as scientific reproducibility and generalisability. We conclude with some reasons for the ineludible importance of interpretability, such as the establishment of trust, in overcoming perhaps the most difficult challenge ML will face in a high-stakes environment like healthcare: professional and public acceptance. (shrink)

The present paper shows how statistical learning theory and machine learning models can be used to enhance understanding of AI-related epistemological issues regarding inductive reasoning and reliability of generalisations. Towards this aim, the paper proceeds as follows. First, it expounds Price’s dual image of representation in terms of the notions of e-representations and i-representations that constitute subject naturalism. For Price, this is not a strictly anti-representationalist position but rather a dualist one (e- and i-representations). Second, the paper (...) links this debate with machine learning in terms of statistical learning theory becoming more viable epistemological tool when it abandons the perspective of object naturalism. The paper then argues that machine learning grounds a form of knowing that can be understood in terms of e- and i-representation learning. Third, this synthesis shows a way of analysing inductive reasoning in terms of reliability of generalisations stemming from a structure of e- and i-representations. In the age of Artificial Intelligence, connecting Price’s dual view of representation with Deep Learning provides an epistemological way forward and even perhaps an approach to how knowing is possible. (shrink)

We provide an analysis of theory-ladenness in machine learning (ML) in science, where ‘theory’ (that we call ‘domain-theory’) refers to the domain knowledge of the scientific discipline where ML is used. By constructing an account of ML models based on a comparison with phenomenological models, we show (against recent trends in philosophy of science) that ML model-building is mostly indifferent to domain-theory, even if the model remains theory-laden in a weak sense, which we call theory-infection. These claims, we (...) argue, have far-reaching consequences for the transferability of ML across scientific disciplines, and shift the priorities of the debate on theory-ladenness in ML from descriptive to normative. (shrink)

The information technological progress that has been achieved over the last decades has also given the humanities the opportunity to expand their methodological toolbox. This paper explores how recent advancements in natural language processing may be used for research in the history of ideas so as to overcome traditional scholarship's inevitably selective approach to historical sources. By employing two machine learning techniques whose potential for the analysis of conceptual continuities and innovations has never been considered before, we aim (...) to determine the extent to which they can enhance conventional research methods. It will amount to a critical evaluation of how the advantages of computational in-breadth analysis could be combined with the merits of traditional in-depth analysis in a philosophically fruitful way. After a brief technical description, the approach will be applied to an example: the conceptual (dis)continuity between medieval and early modern philosophy. All the challenges encountered during development and application will be carefully evaluated. We will then be able to assess whether these tools and techniques present promising extensions to the methodological toolbox of traditional scholarship, or whether they do not yet have the potential for a task as complex as the analysis of philosophical literature. The present investigation can thus be seen as an experiment on how far one can go with current machine-learning techniques in this area of research. In doing so, it provides important insights and guidance for future advances in the field. (shrink)

Learning general concepts in imperfect environments is difficult since training instances often include noisy data, inconclusive data, incomplete data, unknown attributes, unknown attribute values and other barriers to effective learning. It is well known that people can learn effectively in imperfect environments, and can manage to process very large amounts of data. Imitating human learning behavior therefore provides a useful model for machine learning in real-world applications. This paper proposes a new, more effective way to (...) represent imperfect training instances and rules, and based on the new representation, a Human-Like Learning (HULL) algorithm for incrementally learning concepts well in imperfect training environments. Several examples are given to make the algorithm clearer. Finally, experimental results are presented that show the proposed learning algorithm works well in imperfect learning environments. (shrink)

The impact of machine learning in healthcare on patient informed consent is now the subject of significant inquiry in bioethics. However, the topic has predominantly been considered in the context of black box diagnostic or treatment recommendation algorithms. The impact of machine learning involved in healthcare resource allocation on patient consent remains undertheorized. This paper will establish where patient consent is relevant in healthcare resource allocation, before exploring the impact on informed consent from the introduction of (...) black box machine learning into resource allocation. It will then consider the arguments for informing patients about the use of machine learning in resource allocation, before exploring the challenge of whether individual patients could principally contest algorithmic prioritization decisions involving black box machine learning. Finally, this paper will examine how different forms of opacity in machine learning involved in resource allocation could be a barrier to patient consent to clinical decision-making in different healthcare contexts. (shrink)

The workshop “Machine Learning: Prediction Without Explanation?” brought together philosophers of science and scholars from various fields who study and employ Machine Learning (ML) techniques, in order to discuss the changing face of science in the light of ML's constantly growing use. One major focus of the workshop was on the impact of ML on the concept and value of scientific explanation. One may speculate whether ML’s increased use in science exemplifies a paradigmatic turn towards mere (...) pattern recognition and prediction and away from science’s traditional aim of explanation. In contrast, certain conceptions of explanation, such as statistical explanation, could turn out to fit well with the achievements of present-day ML and concede an explanatory value to these achievements after all. It is an open question how to explain ML successes themselves, and this question was in the focus of several talks in the workshop. Based on the topics raised, we will discuss the talks’ contents in more detail as organized into (i) practitioners’ perspectives, (ii) explanations from ML, (iii) explanations of ML, (iv) societal implications, and (v) global and historical perspectives. (shrink)

As more objections have been raised against grant peer-review for being costly and time-consuming, the legitimate question arises whether machine learning algorithms could help assess the epistemic efficiency of the proposed projects. As a case study, we investigated whether project efficiency in high energy physics can be algorithmically predicted based on the data from the proposal. To analyze the potential of algorithmic prediction in HEP, we conducted a study on data about the structure and outcomes of HEP experiments (...) with the goal of predicting their efficiency. In the first step, we assessed the project efficiency using Data Envelopment Analysis of 67 experiments conducted in the HEP laboratory Fermilab. In the second step, we employed predictive algorithms to detect which team structures maximize the epistemic performance of an expert group. For this purpose, we used the efficiency scores obtained by DEA and applied predictive algorithms – lasso and ridge linear regression, neural network, and gradient boosted trees – on them. The results of the predictive analyses show moderately high accuracy, indicating that they can be beneficial as one of the steps in grant review. Still, their applicability in practice should be approached with caution. Some of the limitations of the algorithmic approach are the unreliability of citation patterns, unobservable variables that influence scientific success, and the potential predictability of the model. (shrink)

The increasing integration of Machine Learning (ML) techniques into clinical care, driven in particular by Deep Learning (DL) using Artificial Neural Nets (ANNs), promises to reshape medical practice on various levels and across multiple medical fields. Much recent literature examines the ethical consequences of employing ML within medical and psychiatric practice but the potential impact on psychiatric diagnostic systems has so far not been well-developed. In this article, we aim to explore the challenges that arise from the (...) recent use of ANNs for the old problems of psychiatric nosology. To enable an empirically supported critical reflection on the topic, we conducted semi-structured qualitative interviews with Swiss and German experts in computational psychiatry. Here, we report our findings structured around two themes, namely (1) the possibility of using ML for defining or refining of psychiatric classification, and (2) the desirability of employing ML for psychiatric nosology. We discuss these themes by relating them to recent debates about network theory for psychiatric nosology and show why empirical research in the field should critically reflect on its contribution to psychopathology research. In sum, we argue that beyond technical, regulatory, and ethical challenges, philosophical reflection is crucial to harness the potential of ML in psychiatry. (shrink)

This paper develops an account of how the implementation of ML models into healthcare settings requires revising the methodological apparatus of philosophical bioethics. On this account, ML models are cognitive interventions that provide decision-support to physicians and patients. Due to reliability issues, opaque reasoning processes, and information asymmetries, ML models pose inferential problems for them. These inferential problems lay the grounds for many ethical problems that currently claim centre-stage in the bioethical debate. Accordingly, this paper argues that the best way (...) to make progress in remedying these ethical problems is to distil their epistemic core and to identify appropriate epistemic norms as guardrails. The viability of this approach will be highlighted based on four key issues: trust, responsibility, paternalism, and fairness. In that respect, the paper also contributes to a more pronounced view of the specific methodological challenges for ethical theorizing, when the point of reference are complex statistical models that make predictions, rather than individual agents, acting for reasons. (shrink)

The use of AI algorithms in criminal trials has been the subject of very lively ethical and legal debates recently. While there are concerns over the lack of accuracy and the harmful biases that certain algorithms display, new algorithms seem more promising and might lead to more accurate legal decisions. Algorithms seem especially relevant for bail decisions, because such decisions involve statistical data to which human reasoners struggle to give adequate weight. While getting the right legal outcome is a strong (...) desideratum of criminal trials, advocates of the relational theory of procedural justice give us good reason to think that fairness and perceived fairness of legal procedures have a value that is independent from the outcome. According to this literature, one key aspect of fairness is trustworthiness. In this paper, I argue that using certain algorithms to assist bail decisions could increase three diferent aspects of judges’ trustworthiness: (1) actual trustworthiness, (2) rich trustworthiness, and (3) perceived trustworthiness. (shrink)

Rapid advancements in machine learning techniques allow mass surveillance to be applied on larger scales and utilize more and more personal data. These developments demand reconsideration of the privacy-security dilemma, which describes the tradeoffs between national security interests and individual privacy concerns. By investigating mass surveillance techniques that use bulk data collection and machine learning algorithms, we show why these methods are unlikely to pinpoint terrorists in order to prevent attacks. The diverse characteristics of terrorist attacks—especially (...) when considering lone-wolf terrorism—lead to irregular and isolated footprints. The irregularity of data affects the accuracy of machine learning algorithms and the mass surveillance that depends on them which can be explained by three kinds of known problems encountered in machine learning theory: class imbalance, the curse of dimensionality, and spurious correlations. Proponents of mass surveillance often invoke the distinction between collecting data and metadata, in which the latter is understood as a lesser breach of privacy. Their arguments commonly overlook the ambiguity in the definitions of data and metadata and ignore the ability of machine learning techniques to infer the former from the latter. Given the sparsity of datasets used for machine learning in counterterrorism and the privacy risks attendant with bulk data collection, policymakers and other relevant stakeholders should critically re-evaluate the likelihood of success of the algorithms and the collection of data on which they depend. (shrink)

Machine learning (ML) is increasingly used to interrogate biological systems whose complexity resists law-like, deductive explanation. As a result, embeddings, clusters, and attributions are often overinterpreted, dependencies are left implicit, and claims about explainability are often insufficiently bounded. In this work, we present a framework for contextualizing how machine learning contributes to scientific understanding in biology via compression, qualitative intelligibility, and dependency models. Compression is achieved when inductive biases encode biological structure, reducing the effective hypothesis space (...) and yielding representations aligned with known biology. Qualitative intelligibility is supported when high-dimensional measurements are mapped to human-graspable objects, such as embeddings, clusters, and trajectories, that enable accurate qualitative reasoning without exact calculation. Dependency modelling is realized when learned models make explicit the pattern of relations among system components and thereby guide prediction and intervention. We examine how these principles manifest in successful ML applications and discuss considerations that emerge from this framework. Overall, when viewed through these lenses, ML can transform predictive success into intervention-guiding knowledge in the life sciences. (shrink)

Judgments of misinformation are made relative to the informational preferences of the communities making them. However, informational standards change over time, inducing distribution shifts that threaten the adequacy of machine learning models of misinformation. After articulating five kinds of distribution shifts, three solutions for enhancing success are discussed: larger static training sets, social engineering, and dynamic sampling. I argue that given the idiosyncratic ontology of misinformation, the first option is inadequate, the second is unethical, and thus the third (...) is superior. However, I conclude that the prospects for machine learning models of misinformation are far weaker than most have presupposed, given that both epistemic and non-epistemic values are difficult to operationalize dynamically in machine code, rendering them surprisingly at most a species of recommender systems rather than literal truth detectors. (shrink)

As scientists start to adopt machine learning (ML) as one research tool, the security of ML and the knowledge generated become a concern. In this paper, I explain how supervised ML can be improved with better data ontology, or the way we make categories and turn information into data. More specifically, we should design data ontology in such a way that is consistent with the knowledge that we have about the target phenomenon so that such ontology can help (...) us make the inductive leap. I do so by thinking through a thought experiment, Goodman’s New Riddle of Induction (Fact, fiction, and forecast, Harvard University Press, 1955). Goodman’s riddle helps flesh out three problems of induction: (1) the problem of equal goodies, that there are often too many equally good inductive results given the same data; (2) the problem of diverging performance, that these equally good results can give opposite predictions in the future; and (3) the problem of mediocrity, that when averaged across all equally possible datasets and tasks, no inductive algorithm outperforms any other. I show that all these three problems are manifested as real obstacles in ML practice, namely, the Rashomon effect (Breiman in Stat Sci 16(3):199–231, 2001), the problem of underspecification (D’Amour et al. in J Mach Learn Res, 2020, https://doi.org/10.48550/arXiv.2011.03395 ), and the No Free Lunch theorem (Wolpert in Neural Comput 8(7):1341–90, 1996, https://doi.org/10.1162/neco.1996.8.7.1341 ). Lastly, I argue that proper data ontology can help mitigate these problems and I demonstrate how using concrete examples from climate science. This research highlights the links between philosophers’ discussions of induction and implications in ML practice. (shrink)

This paper establishes a connection between the fields of machine learning (ML) and philosophy concerning the phenomenon of behaving neutrally. It investigates a specific class of ML systems capable of delivering a neutral response to a given task, referred to as abstaining machine learning systems, that has not yet been studied from a philosophical perspective. The paper introduces and explains various abstaining machine learning systems, and categorizes them into distinct types. An examination is conducted (...) on how abstention in the different machine learning system types aligns with the epistemological counterpart of suspended judgment, addressing both the nature of suspension and its normative profile. Additionally, a philosophical analysis is suggested on the autonomy and explainability of the abstaining response. It is argued, specifically, that one of the distinguished types of abstaining systems is preferable as it aligns more closely with our criteria for suspended judgment. Moreover, it is better equipped to autonomously generate abstaining outputs and offer explanations for abstaining outputs when compared to the other type. (shrink)

Along with potential benefits to healthcare delivery, machine learning healthcare applications raise a number of ethical concerns. Ethical evaluations of ML-HCAs will need to structure th...

The increasing precision of observations of the large-scale structure of the universe has created a problem for simulators: running the simulations necessary to interpret these observations has become impractical. Simulators have thus turned to machine learning (ML) algorithms instead. Though ML decreases computational expense, one might be worried about the use of ML for scientific investigations: How can algorithms that have repeatedly been described as black-boxes deliver scientific understanding? In this paper, I investigate how cosmologists employ ML, arguing (...) that in this context, ML algorithms should not be considered black-boxes and can deliver genuine scientific understanding. Accordingly, understanding the methodological role of ML algorithms is crucial to understanding the types of questions they are capable of, and ought to be responsible for, answering. (shrink)

Museum collection databases contain echoes of encounter between colonial collectors (broadly defined) and Indigenous people from around the world. The moment of acquisition—when an item passed out of a community and into the hands of the collector—often included multilingual acts of translation. An artist may have shared the Indigenous name of the object, or the terms associated with its origin and use. Late nineteenth and twemtieth century museum registrars would in turn transcribe this information from field logs into museum catalogs. (...) Over time, these catalog entries were transformed into digital records within collections managements systems (e.g., EMu, PastPerfect, etc.). As a result of this 150-year process, today’s museum collection databases are riddled with Indigenous words and descriptions, scattered across various metadata fields. They may include Native place-names, family names or vocabulary terms that, when translated, extend far beyond the categories ascribed by museum collection managers. These instances of Indigenous description may also serve as a crucial bridge for reconnecting source communities with items of particular interest to their cultural heritage and linguistic preservation efforts. Aiming to enhance the accessibility of Indigenous languages contained in the metadata of cultural heritage collections, this paper explores applications of machine learning methodologies to identify Indigenous terms present in museum catalogs. Specifically, we discuss methods that incorporate the Google Cloud Language Identification Service to detect A:shiwi (Pueblo of Zuni) language terms through a case study of metadata records from the two largest natural history museums in the USA. We utilize an elimination mechanism to exclude specific languages (e.g., English and Spanish) at the word and phrase levels to detect A:shiwi terms. Our approach outperforms conventional methods in terms of accuracy, recall, precision, and F1-scores. This method can be used to confront the “Digital Heap” of cultural heritage records across institutions to improve the discoverability of Indigenous languages in metadata descriptions and reconnect source communities with items of cultural patrimony. (shrink)

In recent years, the work of organizations in the area of digitization has intensified significantly. This trend is also evident in the field of recruitment where job application tracking systems (ATS) have been developed to allow job advertisements to be published online. However, recent studies have shown that recruiting in most organizations is not inclusive, being subject to human biases and prejudices. Most discrimination activities appear early but subtly in the hiring process, for instance, exclusive phrasing in job advertisement discourages (...) qualified applicants from minority groups from applying. The existing works are limited to analyzing, categorizing and highlighting the occurrence of bias in the recruitment process. In this paper, we go beyond this and develop machine learning models for identifying and classifying biased and discriminatory language in job descriptions. We develop and evaluate a machine learning system for identifying five major categories of biased and discriminatory language in job advertisements, i.e., masculine-coded, feminine-coded, exclusive, LGBTQ-coded, demographic and racial language. We utilized the combination of linguistic features with recent state-of-the-art word embeddings representations as input features for various machine learning classifiers. Our results show that the machine learning classifiers were able to identify all the five categories of biased and discriminatory language with a decent accuracy. The Random Forest classifier with FastText word embeddings achieved the best performance with tenfolds cross-validation. Our system directly addresses the bias in the attraction phase of hiring by identifying and classifying biased and discriminatory language and thus encouraging recruiters to write more inclusive job advertisements. (shrink)

Typically, fair machine learning research focuses on a single decision maker and assumes that the underlying population is stationary. However, many of the critical domains motivating this work are characterized by competitive marketplaces with many decision makers. Realistically, we might expect only a subset of them to adopt any non-compulsory fairness-conscious policy, a situation that political philosophers call partial compliance. This possibility raises important questions: how does partial compliance and the consequent strategic behavior of decision subjects affect the (...) allocation outcomes? If k% of employers were to voluntarily adopt a fairness-promoting intervention, should we expect k% progress (in aggregate) towards the benefits of universal adoption, or will the dynamics of partial compliance wash out the hoped-for benefits? How might adopting a global (versus local) perspective impact the conclusions of an auditor? In this paper, we propose a simple model of an employment market, leveraging simulation as a tool to explore the impact of both interaction effects and incentive effects on outcomes and auditing metrics. Our key findings are that at equilibrium: (1) partial compliance by k% of employers can result in far less than proportional (k%) progress towards the full compliance outcomes; (2) the gap is more severe when fair employers match global (vs local) statistics; (3) choices of local vs global statistics can paint dramatically different pictures of the performance vis-a-vis fairness desiderata of compliant versus non-compliant employers; (4) partial compliance based on local parity measures can induce extreme segregation. Finally, we discuss implications for auditors and insights concerning the design of regulatory frameworks. (shrink)

Introduction In their contribution, Ugar and Malele 1 shed light on an often overlooked but crucial aspect of the ethical development of machine learning (ML) systems to support the diagnosis of mental health disorders. The authors restrain their focus on pointing to the danger of misdiagnosing mental health pathologies that do not qualify as such within sub-Saharan African communities and argue for the need to include population-specific values in these technologies’ design. However, an analysis of the nature of (...) the harm caused to said populations once their values remain unrecognised is not offered. Building on Ugar and Malele’s considerations, we add a further perspective to their analysis by showing the need to design considering intended values to avoid the occurrence of epistemic injustices. 2 First, we argue that failing to acknowledge the hermeneutical offerings of the populations interacting with these systems can qualify as _contributory injustice_.shrink)