-
Explanations of Black-Box Models based on Directional Feature Interactions
Authors:
Aria Masoomi,
Davin Hill,
Zhonghui Xu,
Craig P Hersh,
Edwin K. Silverman,
Peter J. Castaldi,
Stratis Ioannidis,
Jennifer Dy
Abstract:
As machine learning algorithms are deployed ubiquitously to a variety of domains, it is imperative to make these often black-box models transparent. Several recent works explain black-box models by capturing the most influential features for prediction per instance; such explanation methods are univariate, as they characterize importance per feature. We extend univariate explanation to a higher-or…
▽ More
As machine learning algorithms are deployed ubiquitously to a variety of domains, it is imperative to make these often black-box models transparent. Several recent works explain black-box models by capturing the most influential features for prediction per instance; such explanation methods are univariate, as they characterize importance per feature. We extend univariate explanation to a higher-order; this enhances explainability, as bivariate methods can capture feature interactions in black-box models, represented as a directed graph. Analyzing this graph enables us to discover groups of features that are equally important (i.e., interchangeable), while the notion of directionality allows us to identify the most influential features. We apply our bivariate method on Shapley value explanations, and experimentally demonstrate the ability of directional explanations to discover feature interactions. We show the superiority of our method against state-of-the-art on CIFAR10, IMDB, Census, Divorce, Drug, and gene data.
△ Less
Submitted 15 April, 2023;
originally announced April 2023.
-
Using Machine Learning to Emulate Agent-Based Simulations
Authors:
Claudio Angione,
Eric Silverman,
Elisabeth Yaneske
Abstract:
In this proof-of-concept work, we evaluate the performance of multiple machine-learning methods as statistical emulators for use in the analysis of agent-based models (ABMs). Analysing ABM outputs can be challenging, as the relationships between input parameters can be non-linear or even chaotic even in relatively simple models, and each model run can require significant CPU time. Statistical emul…
▽ More
In this proof-of-concept work, we evaluate the performance of multiple machine-learning methods as statistical emulators for use in the analysis of agent-based models (ABMs). Analysing ABM outputs can be challenging, as the relationships between input parameters can be non-linear or even chaotic even in relatively simple models, and each model run can require significant CPU time. Statistical emulation, in which a statistical model of the ABM is constructed to facilitate detailed model analyses, has been proposed as an alternative to computationally costly Monte Carlo methods. Here we compare multiple machine-learning methods for ABM emulation in order to determine the approaches best suited to emulating the complex behaviour of ABMs. Our results suggest that, in most scenarios, artificial neural networks (ANNs) and gradient-boosted trees outperform Gaussian process emulators, currently the most commonly used method for the emulation of complex computational models. ANNs produced the most accurate model replications in scenarios with high numbers of model runs, although training times were longer than the other methods. We propose that agent-based modelling would benefit from using machine-learning methods for emulation, as this can facilitate more robust sensitivity analyses for the models while also reducing CPU time consumption when calibrating and analysing the simulation.
△ Less
Submitted 24 July, 2021; v1 submitted 5 May, 2020;
originally announced May 2020.
-
Social and Child Care Provision in Kinship Networks: an Agent-Based Model
Authors:
Umberto Gostoli,
Eric Silverman
Abstract:
Providing for the needs of the vulnerable is a critical component of social and health policy-making. In particular, caring for children and for vulnerable older people is vital to the wellbeing of millions of families throughout the world. In most developed countries, this care is provided through both formal and informal means, and is therefore governed by complex policies that interact in non-o…
▽ More
Providing for the needs of the vulnerable is a critical component of social and health policy-making. In particular, caring for children and for vulnerable older people is vital to the wellbeing of millions of families throughout the world. In most developed countries, this care is provided through both formal and informal means, and is therefore governed by complex policies that interact in non-obvious ways with other areas of policy-making. In this paper we present an agent-based model of social and child care provision in the UK, in which agents can provide informal care or pay for private care for their relatives. Agents make care decisions based on numerous factors including their health status, employment, financial situation, and social and physical distance to those in need. Simulation results show that the model can produce plausible patterns of care need and availability, and therefore can provide an important aid to this complex area of policy-making. We conclude that the model's use of kinship networks for distributing care and the explicit modelling of interactions between social care and child care will enable policy-makers to develop more informed policy interventions in these critical areas.
△ Less
Submitted 28 February, 2020; v1 submitted 12 February, 2020;
originally announced February 2020.
-
Situating Agent-Based Modelling in Population Health Research
Authors:
Eric Silverman,
Umberto Gostoli,
Stefano Picascia,
Jonatan Almagor,
Mark McCann,
Richard Shaw,
Claudio Angione
Abstract:
Today's most troublesome population health challenges are often driven by social and environmental determinants, which are difficult to model using traditional epidemiological methods. We agree with those who have argued for the wider adoption of agent-based modelling (ABM) in taking on these challenges. However, while ABM has been used occasionally in population health, we argue that for ABM to b…
▽ More
Today's most troublesome population health challenges are often driven by social and environmental determinants, which are difficult to model using traditional epidemiological methods. We agree with those who have argued for the wider adoption of agent-based modelling (ABM) in taking on these challenges. However, while ABM has been used occasionally in population health, we argue that for ABM to be most effective in the field it should be used as a means for answering questions normally inaccessible to the traditional epidemiological toolkit. In an effort to clearly illustrate the utility of ABM for population health research, and to clear up persistent misunderstandings regarding the method's conceptual underpinnings, we offer a detailed presentation of the core concepts of complex systems theory, and summarise why simulations are essential to the study of complex systems. We then examine the current state of the art in ABM for population health, and propose they are well-suited for the study of the `wicked' problems in population health, and could make significant contributions to theory and intervention development in these areas.
△ Less
Submitted 6 February, 2020;
originally announced February 2020.
-
Status in flux: Unequal alliances can create power vacuums
Authors:
John Bryden,
Eric Silverman,
Simon T. Powers
Abstract:
Human groups show a variety of leadership structures from no leader, to changing leaders, to a single long-term leader. When a leader is deposed, the presence of a power vacuum can mean they are often quickly replaced. We lack an explanation of how such phenomena can emerge from simple rules of interaction between individuals. Here, we model transitions between different phases of leadership struc…
▽ More
Human groups show a variety of leadership structures from no leader, to changing leaders, to a single long-term leader. When a leader is deposed, the presence of a power vacuum can mean they are often quickly replaced. We lack an explanation of how such phenomena can emerge from simple rules of interaction between individuals. Here, we model transitions between different phases of leadership structure. We find a novel class of group dynamical behaviour where there is a single leader who is quickly replaced when they lose status, demonstrating a power vacuum. The model uses a dynamic network of individuals who non-coercively form and break alliances with one-another, with a key parameter modelling inequality in these alliances. We argue the model can explain transitions in leadership structure in the Neolithic Era from relatively equal hunter-gatherer societies, to groups with chieftains which change over time, to groups with an institutionalised leader on a paternal lineage. Our model demonstrates how these transitions can be explained by the impact of technological developments such as food storage and/or weapons, which meant that alliances became more unequal. In general terms, our approach provides a quantitative understanding of how technology and social norms can affect leadership dynamics and structures.
△ Less
Submitted 24 October, 2019; v1 submitted 4 September, 2019;
originally announced September 2019.
-
Modelling Social Care Provision in An Agent-Based Framework with Kinship Networks
Authors:
Umberto Gostoli,
Eric Silverman
Abstract:
Current demographic trends in the UK include a fast-growing elderly population and drop** birth rates, and demand for social care amongst the aged is rising. The UK depends on informal social care -- family members or friends providing care -- for some 50\% of care provision. However, lower birth rates and a graying population mean that care availability is becoming a significant problem, causin…
▽ More
Current demographic trends in the UK include a fast-growing elderly population and drop** birth rates, and demand for social care amongst the aged is rising. The UK depends on informal social care -- family members or friends providing care -- for some 50\% of care provision. However, lower birth rates and a graying population mean that care availability is becoming a significant problem, causing concern amongst policy-makers that substantial public investment in formal care will be required in decades to come. In this paper we present an agent-based simulation of care provision in the UK, in which individual agents can decide to provide informal care, or pay for private care, for their loved ones. Agents base these decisions on factors including their own health, employment status, financial resources, relationship to the individual in need, and geographical location. Results demonstrate that the model can produce similar patterns of care need and availability as is observed in the real world, despite the model containing minimal empirical data. We propose that our model better captures the complexities of social care provision than other methods, due to the socioeconomic details present and the use of kinship networks to distribute care amongst family members.
△ Less
Submitted 10 April, 2019;
originally announced April 2019.
