Solving the Black Box Problem: A Normative Framework for Explainable Artificial Intelligence

Zednik, Carlos

doi:10.1007/s13347-019-00382-7

Cited by 208 publications

(157 citation statements)

References 39 publications

Supporting

Mentioning

120

Contrasting

Unclassified

Order By: Relevance

“…These additional selection methods may minimize the loss of features that could be influenced by human bias. The black box problem of NN models also makes it challenging to confirm which features were selected into the model [55]. Lastly, unlike past studies that utilized raw signal data captured by IMU sensors [28][29][30][31], the current study utilized pre-processed data (e.g., gait speed, sway area, cadence, etc.)…”

Section: Discussionmentioning

confidence: 99%

Classification of Parkinson’s disease and essential tremor based on balance and gait characteristics from wearable motion sensors via machine learning techniques: a data-driven approach

Moon

Song

Sharma

et al. 2020

J NeuroEngineering Rehabil

View full text Add to dashboard Cite

Background Parkinson’s disease (PD) and essential tremor (ET) are movement disorders that can have similar clinical characteristics including tremor and gait difficulty. These disorders can be misdiagnosed leading to delay in appropriate treatment. The aim of the study was to determine whether balance and gait variables obtained with wearable inertial motion sensors can be utilized to differentiate between PD and ET using machine learning. Additionally, we compared classification performances of several machine learning models. Methods This retrospective study included balance and gait variables collected during the instrumented stand and walk test from people with PD (n = 524) and with ET (n = 43). Performance of several machine learning techniques including neural networks, support vector machine, k-nearest neighbor, decision tree, random forest, and gradient boosting, were compared with a dummy model or logistic regression using F1-scores. Results Machine learning models classified PD and ET based on balance and gait characteristics better than the dummy model (F1-score = 0.48) or logistic regression (F1-score = 0.53). The highest F1-score was 0.61 of neural network, followed by 0.59 of gradient boosting, 0.56 of random forest, 0.55 of support vector machine, 0.53 of decision tree, and 0.49 of k-nearest neighbor. Conclusions This study demonstrated the utility of machine learning models to classify different movement disorders based on balance and gait characteristics collected from wearable sensors. Future studies using a well-balanced data set are needed to confirm the potential clinical utility of machine learning models to discern between PD and ET.

show abstract

Section: Discussionmentioning

confidence: 99%

Classification of Parkinson’s disease and essential tremor based on balance and gait characteristics from wearable motion sensors via machine learning techniques: a data-driven approach

Moon

Song

Sharma

et al. 2020

J NeuroEngineering Rehabil

View full text Add to dashboard Cite

show abstract

“…2 However, though a lot of interesting work has been done in the area of transparency, far less attention has been devoted to the role of transparency in terms of how those who are ultimately affected (i.e., the general public) come to perceive AI decision-making as being legitimate and worthy of acceptance. Researchers have noted the importance of public acceptance with regard to AI implementation (e.g., Zerilli et al 2018;Binns et al 2018) and there are several frameworks that can be used to make AI systems less biased, more fair, etc., (e.g., Binns 2018;Boscoe 2019;Zednik 2019) which might lead to an increase in perceived legitimacy. These frameworks etc., however, do not explicitly engage with the theories and empirical findings from the social sciences regarding how individuals' legitimacy perceptions are affected by different elements, such as the transparency of the process, decisions, or reasons behind said decisions.…”

Section: Introductionmentioning

confidence: 99%

Artificial intelligence, transparency, and public decision-making

Licht

2020

AI & Soc

142

View full text Add to dashboard Cite

The increasing use of Artificial Intelligence (AI) for making decisions in public affairs has sparked a lively debate on the benefits and potential harms of self-learning technologies, ranging from the hopes of fully informed and objectively taken decisions to fear for the destruction of mankind. To prevent the negative outcomes and to achieve accountable systems, many have argued that we need to open up the "black box" of AI decision-making and make it more transparent. Whereas this debate has primarily focused on how transparency can secure high-quality, fair, and reliable decisions, far less attention has been devoted to the role of transparency when it comes to how the general public come to perceive AI decision-making as legitimate and worthy of acceptance. Since relying on coercion is not only normatively problematic but also costly and highly inefficient, perceived legitimacy is fundamental to the democratic system. This paper discusses how transparency in and about AI decision-making can affect the public's perception of the legitimacy of decisions and decision-makers and produce a framework for analyzing these questions. We argue that a limited form of transparency that focuses on providing justifications for decisions has the potential to provide sufficient ground for perceived legitimacy without producing the harms full transparency would bring.

show abstract

“…Several philosophers have analyzed these concepts more carefully. Zednik (2019), for instance, offers a pragmatic account of opacity to provide a normative framework detailing different kinds of knowledge that ought to be required by different stakeholders. Explanations, on this view, are the basis for how such knowledge is acquired.…”

Section: Introductionmentioning

confidence: 99%

What is Interpretability?

2020

View full text Add to dashboard Cite

We argue that artificial networks are explainable and offer a novel theory of interpretability. Two sets of conceptual questions are prominent in theoretical engagements with artificial neural networks, especially in the context of medical artificial intelligence: (1) Are networks explainable, and if so, what does it mean to explain the output of a network? And (2) what does it mean for a network to be interpretable? We argue that accounts of “explanation” tailored specifically to neural networks have ineffectively reinvented the wheel. In response to (1), we show how four familiar accounts of explanation apply to neural networks as they would to any scientific phenomenon. We diagnose the confusion about explaining neural networks within the machine learning literature as an equivocation on “explainability,” “understandability” and “interpretability.” To remedy this, we distinguish between these notions, and answer (2) by offering a theory and typology of interpretation in machine learning. Interpretation is something one does to an explanation with the aim of producing another, more understandable, explanation. As with explanation, there are various concepts and methods involved in interpretation: Total or Partial, Global or Local, and Approximative or Isomorphic. Our account of “interpretability” is consistent with uses in the machine learning literature, in keeping with the philosophy of explanation and understanding, and pays special attention to medical artificial intelligence systems.

show abstract

Solving the Black Box Problem: A Normative Framework for Explainable Artificial Intelligence

Cited by 208 publications

References 39 publications

Classification of Parkinson’s disease and essential tremor based on balance and gait characteristics from wearable motion sensors via machine learning techniques: a data-driven approach

Classification of Parkinson’s disease and essential tremor based on balance and gait characteristics from wearable motion sensors via machine learning techniques: a data-driven approach

Artificial intelligence, transparency, and public decision-making

What is Interpretability?

Contact Info

Product

Resources

About