DFDaniel Fryer

Daniel Fryer

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Daniel Fryer has a total of 66 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Multi-choice Solutions for Feature and Parameter Importance

    by Daniel Fryer, David Lowing, Inga Strümke, Hien Nguyen


  • The wealth of features: towards a coherent cooperative game theory for feature importance

    by Daniel Fryer

    The University of Queensland's institutional repository, UQ eSpace, aims to create global visibility and accessibility of UQ’s scholarly research.

    • 24 October 2023
    • Published by HAL in HAL
    Open Access

    Multi-choice Explanations: A New Cooperative Game Structure for XAI

    by Daniel Fryer, David Lowing, Inga Strümke, Hien Nguyen

    Cooperative game theorists propose the following attractive process: (1) capture the abstract value of each possible coalition of individuals, (2) write down some principles, or axioms, on how to distribute the value (e.g., allocate importance to features or parameters), and then, (3) find a set of allocations that satisfy the principles. The Shapley value has received much attention -- but it is just one solution concept, satisfying one set of principles, in one class of games. It is popular among game theorists because the axioms, and the class of TU-games, are reasonable in game theory. In AI and ML, we should choose carefully what is reasonable for our own purposes. In this paper, we highlight solution concepts in the class of multi-choice games (MC-games). These are model agnostic, and unique to their own set of axioms, just like the Shapley value. This paper offers a general algorithm for constructing any MC-game framework with polynomial time complexity in the number of parameter levels, and an application of this algorithm that is transparent, and can be readily generalised to local explanation frameworks such as SHapley Additive exPlanations (SHAP).

  • Top Cited Papers
    Open Access

    ISARIC-COVID-19 dataset: A Prospective, Standardized, Global Dataset of Patients Hospitalized with COVID-19

    by Esteban Garcia-Gallo, Laura Merson, Kalynn Kennon, Sadie Kelly, Barbara Wanjiru Citarella, Daniel Vidali Fryer, Sally Shrapnel, James Lee, Sara Duque, Yuli V. Fuentes, et al.

    The International Severe Acute Respiratory and Emerging Infection Consortium (ISARIC) COVID-19 dataset is one of the largest international databases of prospectively collected clinical data on people hospitalized with COVID-19. This dataset was compiled during the COVID-19 pandemic by a network of hospitals that collect data using the ISARIC-World Health Organization Clinical Characterization Protocol and data tools. The database includes data from more than 705,000 patients, collected in more than 60 countries and 1,500 centres worldwide. Patient data are available from acute hospital admissions with COVID-19 and outpatient follow-ups. The data include signs and symptoms, pre-existing comorbidities, vital signs, chronic and acute treatments, complications, dates of hospitalization and discharge, mortality, viral strains, vaccination status, and other data. Here, we present the dataset characteristics, explain its architecture and how to gain access, and provide tools to facilitate its use.


  • Open Access

    Use of an extended KDIGO definition to diagnose acute kidney injury in patients with COVID-19: A multinational study using the ISARIC–WHO clinical characterisation protocol

    by Marina Wainstein, Samual MacDonald, Daniel Fryer, Kyle Young, Valeria Balan, Husna Begum, Aidan Burrell, Barbara Wanjiru Citarella, J. Perren Cobb, Sadie Kelly, et al.

    Background: Acute kidney injury (AKI) is one of the most common and significant problems in patients with Coronavirus Disease 2019 (COVID-19). However, little is known about the incidence and impact of AKI occurring in the community or early in the hospital admission. The traditional Kidney Disease Improving Global Outcomes (KDIGO) definition can fail to identify patients for whom hospitalisation coincides with recovery of AKI as manifested by a decrease in serum creatinine (sCr). We hypothesised that an extended KDIGO (eKDIGO) definition, adapted from the International Society of Nephrology (ISN) 0by25 studies, would identify more cases of AKI in patients with COVID-19 and that these may correspond to community-acquired AKI (CA-AKI) with similarly poor outcomes as previously reported in this population. Methods and findings: All individuals recruited using the International Severe Acute Respiratory and Emerging Infection Consortium (ISARIC)–World Health Organization (WHO) Clinical Characterisation Protocol (CCP) and admitted to 1,609 hospitals in 54 countries with Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) infection from February 15, 2020 to February 1, 2021 were included in the study. Data were collected and analysed for the duration of a patient’s admission. Incidence, staging, and timing of AKI were evaluated using a traditional and eKDIGO definition, which incorporated a commensurate decrease in sCr. Patients within eKDIGO diagnosed with AKI by a decrease in sCr were labelled as deKDIGO. Clinical characteristics and outcomes—intensive care unit (ICU) admission, invasive mechanical ventilation, and in-hospital death—were compared for all 3 groups of patients. The relationship between eKDIGO AKI and in-hospital death was assessed using survival curves and logistic regression, adjusting for disease severity and AKI susceptibility. A total of 75,670 patients were included in the final analysis cohort. Median length of admission was 12 days (interquartile range [IQR] 7, 20). There were twice as many patients with AKI identified by eKDIGO than KDIGO (31.7% versus 16.8%). Those in the eKDIGO group had a greater proportion of stage 1 AKI (58% versus 36% in KDIGO patients). Peak AKI occurred early in the admission more frequently among eKDIGO than KDIGO patients. Compared to those without AKI, patients in the eKDIGO group had worse renal function on admission, more in-hospital complications, higher rates of ICU admission (54% versus 23%) invasive ventilation (45% versus 15%), and increased mortality (38% versus 19%). Patients in the eKDIGO group had a higher risk of in-hospital death than those without AKI (adjusted odds ratio: 1.78, 95% confidence interval: 1.71 to 1.80, p-value < 0.001). Mortality and rate of ICU admission were lower among deKDIGO than KDIGO patients (25% versus 50% death and 35% versus 70% ICU admission) but significantly higher when compared to patients with no AKI (25% versus 19% death and 35% versus 23% ICU admission) (all p-values −5). Limitations include ad hoc sCr sampling, exclusion of patients with less than two sCr measurements, and limited availability of sCr measurements prior to initiation of acute dialysis. Conclusions: An extended KDIGO definition of AKI resulted in a significantly higher detection rate in this population. These additional cases of AKI occurred early in the hospital admission and were associated with worse outcomes compared to patients without AKI.


  • Open Access

    Use of an extended KDIGO definition to diagnose acute kidney injury in patients with COVID-19: A multinational study of the ISARIC cohort

    by Marina Wainstein, Samual MacDonald, Daniel Fryer, Kyle Young, Steven Webb, Husna Begum, Alistair Nichol, James Lee, Valeria Balan, J. Perren Cobb, et al.

    Background: Acute kidney injury (AKI) is one of the most common and significant problems in patients with COVID-19. However, little is known about the incidence and impact of AKI occurring in the community or early in the hospital admission. The traditional KDIGO definition can fail to identify patients for whom hospitalization coincides with recovery of AKI as manifested by a decrease in serum creatinine (sCr). We hypothesized that an extended KDIGO definition, adapted from the International Society of Nephrology 0by25 studies, would identify more cases of AKI in patients with COVID-19 and that these may correspond to community-acquired AKI with similarly poor outcomes as previously reported in this population. Methods and Findings: All individuals in the ISARIC cohort admitted to hospital with SARS-CoV-2 infection from February 15 th , 2020, to February 1 st , 2021, were included in the study. Data was collected and analysed for the duration of a patient’s admission. Incidence, staging and timing of AKI were evaluated using a traditional and extended KDIGO (eKDIGO) definition which incorporated a commensurate decrease in serum creatinine. Patients within eKDIGO diagnosed with AKI by a decrease in sCr were labelled as deKDIGO. Clinical characteristic and outcomes – intensive care unit (ICU) admission, invasive mechanical ventilation and in-hospital death - were compared for all three groups of patients. The relationship between eKDIGO AKI and in-hospital death was assessed using survival curves and logistic regression, adjusting for disease severity and AKI susceptibility. 75,670 patients from 54 countries were included in the final analysis cohort. Median length of admission was 12 days (IQR 7, 20). There were twice as many patients with AKI identified by eKDIGO than KDIGO (31.7 vs 16.8%). Those in the eKDIGO group had a greater proportion of stage 1 AKI (58% vs 36% in KDIGO patients). Peak AKI occurred early in the admission more frequently among eKDIGO than KDIGO patients. Compared to those without AKI, patients in the eKDIGO group had worse renal function on admission, more in-hospital complications, higher rates of ICU admission (54% vs 23%) invasive ventilation (45% vs 15%) and increased mortality (38% vs 19%). Patients in the eKDIGO group had a higher risk of in-hospital death than those without AKI (adjusted OR: 1.78, 95% confidence interval: 1.71-1.8, p-value < 0.001). Mortality and rate of ICU admission were lower among deKDIGO than KDIGO patients (25% vs 50% death and 35% vs 70% ICU admission) but significantly higher when compared to patients with no AKI (25% vs 19% death and 35% vs 23% ICU admission) (all p values < 5×10 −5 ). Limitations include ad hoc sCr sampling, exclusion of patients with less than two sCr measurements, and limited availability of sCr measurements prior to initiation of acute dialysis. Conclusions: The use of an extended KDIGO definition to diagnose AKI in this population resulted in a significantly higher incidence rate compared to traditional KDIGO criteria. These additional cases of AKI appear to be occurring in the community or early in the hospital admission and are associated with worse outcomes than those without AKI. Author Summary: Why was this study done?: Previous studies have shown that acute kidney injury (AKI) is a common problem among hospitalized patients with COVID-19. The current biochemical criteria used to diagnose AKI may be insufficient to capture AKI that develops in the community and is recovering by the time a patient presents to hospital. The use of an extended definition, that can identify AKI both during its development and recovery phase, may allow us to identify more patients with AKI. These patients may benefit from early management strategies to improve long term outcomes. What did the researchers do and find?: In this study, we examined AKI incidence, severity and outcomes among a large international cohort of patients with COVID-19 using both a traditional and extended definition of AKI. We found that using the extended definition identified almost twice as many cases of AKI than the traditional definition (31.7 vs 16.8%). These additional cases of AKI were generally less severe and occurred earlier in the hospital admission. Nevertheless, they were associated with worse outcomes, including ICU admission and in-hospital death (adjusted odds ratio: 1.78, 95% confidence interval: 1.71-1.8, p-value < 0.001) than those with no AKI. What do these findings mean?: The current definition of AKI fails to identify a large group of patients with AKI that appears to develop in the community or early in the hospital admission. Given the finding that these cases of AKI are associated with worse admission outcomes than those without AKI, identifying and managing them in a timely manner is enormously important.


  • Top Cited Papers
    Open Access

    Shapley Values for Feature Selection: The Good, the Bad, and the Axioms

    by Daniel Fryer, Inga Strumke, Hien Nguyen

    The Shapley value has become popular in the Explainable AI (XAI) literature, thanks, to a large extent, to a solid theoretical foundation, including four “favourable and fair” axioms for attribution in transferable utility games. The Shapley value is provably the only solution concept satisfying these axioms. In this paper, we introduce the Shapley value and draw attention to its recent uses as a feature selection tool. We call into question this use of the Shapley value, using simple, abstract “toy” counterexamples to illustrate that the axioms may work against the goals of feature selection. From this, we develop a number of insights that are then investigated in concrete simulation settings, with a variety of Shapley value formulations, including SHapley Additive exPlanations (SHAP) and Shapley Additive Global importancE (SAGE). The aim is not to encourage any use of the Shapley value for feature selection, but we aim to clarify various limitations around their current use in the literature. In so doing, we hope to help demystify certain aspects of the Shapley value axioms that are viewed as “favourable”. In particular, we wish to highlight that the favourability of the axioms depends non-trivially on the way in which the Shapley value is appropriated in the XAI application.


  • Open Access

    Explaining the data or explaining a model? Shapley values that uncover non-linear dependencies

    by Daniel Vidali Fryer, Inga Strümke, Hien Nguyen

    Shapley values have become increasingly popular in the machine learning literature thanks to their attractive axiomatisation, flexibility, and uniqueness in satisfying certain notions of `fairness'. The flexibility arises from the myriad potential forms of the Shapley value \textit{game formulation}. Amongst the consequences of this flexibility is that there are now many types of Shapley values being discussed, with such variety being a source of potential misunderstanding. To the best of our knowledge, all existing game formulations in the machine learning and statistics literature fall into a category which we name the model-dependent category of game formulations. In this work, we consider an alternative and novel formulation which leads to the first instance of what we call model-independent Shapley values. These Shapley values use a (non-parametric) measure of non-linear dependence as the characteristic function. The strength of these Shapley values is in their ability to uncover and attribute non-linear dependencies amongst features. We introduce and demonstrate the use of the energy distance correlations, affine-invariant distance correlation, and Hilbert-Shmidt independence criterion as Shapley value characteristic functions. In particular, we demonstrate their potential value for exploratory data analysis and model diagnostics. We conclude with an interesting expository application to a classical medical survey data set.


  • Open Access

    Shapley Value Confidence Intervals for Attributing Variance Explained

    by Daniel Fryer, Inga Strümke, Hien Nguyen

    The coefficient of determination, the R2, is often used to measure the variance explained by an affine combination of multiple explanatory covariates. An attribution of this explanatory contribution to each of the individual covariates is often sought in order to draw inference regarding the importance of each covariate with respect to the response phenomenon. A recent method for ascertaining such an attribution is via the game theoretic Shapley value decomposition of the coefficient of determination. Such a decomposition has the desirable efficiency, monotonicity, and equal treatment properties. Under a weak assumption that the joint distribution is pseudo-elliptical, we obtain the asymptotic normality of the Shapley values. We then utilize this result in order to construct confidence intervals and hypothesis tests for Shapley values. Monte Carlo studies regarding our results are provided. We found that our asymptotic confidence intervals required less computational time to competing bootstrap methods and are able to exhibit improved coverage, especially on small samples. In an expository application to Australian real estate price modeling, we employ Shapley value confidence intervals to identify significant differences between the explanatory contributions of covariates, between models, which otherwise share approximately the same R2 value. These different models are based on real estate data from the same periods in 2019 and 2020, the latter covering the early stages of the arrival of the novel coronavirus, COVID-19.


  • k-means on Positive Definite Matrices, and an Application to Clustering in Radar Image Sequences

    by Daniel Fryer, Hien Nguyen, Pascal Castellazzi

    We state theoretical properties for k-means clustering of Symmetric Positive Definite (SPD) matrices, in a non-Euclidean space, that provides a natural and favourable representation of these data. We then provide a novel application for this method, to time-series clustering of pixels in a sequence of Synthetic Aperture Radar images, via their finite-lag autocovariance matrices.