Estimation of Counterfactual Interventions under Uncertainties

Parallel Problem Solving from Nature – PPSN XVI

Counterfactual explanations are one of the most popular methods to make predictions of black box machine learning models interpretable by providing explanations in the form of 'what-if scenarios'. Most current approaches optimize a collapsed, weighted sum of multiple objectives, which are naturally difficult to balance a-priori. We propose the Multi-Objective Counterfactuals (MOC) method, which translates the counterfactual search into a multi-objective optimization problem. Our approach not only returns a diverse set of counterfactuals with different trade-offs between the proposed objectives, but also maintains diversity in feature space. This enables a more detailed post-hoc analysis to facilitate better understanding and also more options for actionable user responses to change the predicted outcome. Our approach is also model-agnostic and works for numerical and categorical input features. We show the usefulness of MOC in concrete cases and compare our approach with state-of-the-art methods for counterfactual explanations.

IEEE Access

Increasingly in recent times, the mere prediction of a machine learning algorithm is considered insufficient to gain complete control over the event being predicted. A machine learning algorithm should be considered reliable in the way it allows to extract more knowledge and information than just having a prediction at hand. In this perspective, the counterfactual theory plays a central role. By definition, a counterfactual is the smallest variation of the input such that it changes the predicted behaviour. The paper addresses counterfactuals through Support Vector Data Description (SVDD), empowered by explainability and metric for assessing the counterfactual quality. After showing the specific case in which an analytical solution may be found (under Euclidean distance and linear kernel), an optimisation problem is posed for any type of distances and kernels. The vehicle platooning application is the use case considered to demonstrate how the outlined methodology may offer support to safety-critical applications as well as how explanation may shed new light into the control of the system at hand.

arXiv (Cornell University), 2022

Counterfactuals are central in causal human reasoning and the scientific discovery process. The uplift, also called conditional average treatment effect, measures the causal effect of some action, or treatment, on the outcome of an individual. This paper discusses how it is possible to derive bounds on the probability of counterfactual statements based on uplift terms. First, we derive some original bounds on the probability of counterfactuals and we show that tightness of such bounds depends on the information of the feature set on the uplift term. Then, we propose a point estimator based on the assumption of conditional independence between the counterfactual outcomes. The quality of the bounds and the point estimators are assessed on synthetic data and a large real-world customer data set provided by a telecom company, showing significant improvement over the state of the art.

Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence, 2018

Models based on rules that express local and heterogeneous mechanisms of stochastic interactions between structured agents are an important tool for investigating the dynamical behavior of complex systems, especially in molecular biology. Given a simulated trace of events, the challenge is to construct a causal diagram that explains how a phenomenon of interest occurred. Counterfactual analysis can provide distinctive insights, but its standard definition is not applicable in rule-based models because they are not readily expressible in terms of structural equations. We provide a semantics of counterfactual statements that addresses this challenge by sampling counterfactual trajectories that are probabilistically as close to the factual trace as a given intervention permits them to be. We then show how counterfactual dependencies give rise to explanations in terms of relations of enablement and prevention between events.

The International FLAIRS Conference Proceedings

Counterfactuals have become a useful tool for explainable Artificial Intelligence (XAI). Counterfactuals provide various perturbations to a data instance to yield an alternate classification from a machine learning model. Several algorithms have been designed to generate counterfactuals using deep neural networks; however, despite their growing use in many mission-critical fields, there has been no investigation to date as to the epistemic uncertainty of generated counterfactuals. This could result in the use of risk-prone explanations in these fields. In this work, we use several data sets to compare the epistemic uncertainty of original instances to that of counterfactuals generated from those instances. As part of our analysis, we also measure the extent to which counterfactuals can be considered anomalies in those data sets. We find that counterfactual uncertainty is higher in three of the four datasets tested. Moreover, our experiments suggest a possible connection between reco...

Social Science Research Network, 2020

Karni and Vierø (2013) propose a model of belief revision under growing awarenessreverse Bayesianism-which posits that as a person becomes aware of new acts, consequences, or act-consequence links, she revises her beliefs over an expanded state space in a way that preserves the relative likelihoods of events in the original state space. A key limitation of the model is that reverse Bayesianism alone does not fully determine the revised probability distribution. We provide an assumption-act independencethat imposes additional restrictions on reverse Bayesian belief revision. We show that under act independence, knowledge of the probabilities of new events in the expanded state space is su¢ cient to fully determine the revised probability distribution in each case of growing awareness. We thereby operationalize the reverse Bayesian model for applications. To illustrate how act independence operationalizes reverse Bayesianism, we consider the law and economics problem of optimal safety regulation. JEL codes: D83, K23.

arXiv (Cornell University), 2020

Counterfactual reasoning from logged data has become increasingly important for many applications such as web advertising or healthcare. In this paper, we address the problem of learning stochastic policies with continuous actions from the viewpoint of counterfactual risk minimization (CRM). While the CRM framework is appealing and well studied for discrete actions, the continuous action case raises new challenges about modelization, optimization, and offline model selection with real data which turns out to be particularly challenging. Our paper contributes to these three aspects of the CRM estimation pipeline. First, we introduce a modelling strategy based on a joint kernel embedding of contexts and actions, which overcomes the shortcomings of previous discretization approaches. Second, we empirically show that the optimization aspect of counterfactual learning is important, and we demonstrate the benefits of proximal point algorithms and differentiable estimators. Finally, we propose an evaluation protocol for offline policies in real-world logged systems, which is challenging since policies cannot be replayed on test data, and we release a new large-scale dataset along with multiple synthetic, yet realistic, evaluation setups. 1

IEEE transactions on artificial intelligence, 2022

Explainability has become crucial in Artificial Intelligence studies and, as the complexity of the model increases, so does the complexity of its explanation. However, the higher the complexity of the problem, the higher the amount of information it may provide, and this information can be exploited to generate a more precise explanation of how the model works. One of the most valuable ways to recover such input-output relation is to extract counterfactual explanations that allow us to find minimal changes from an observation to another one belonging to a different class. In this article, we propose a novel methodology to extract multiple counterfactual explanations (MUCH, MUlti Counterfactual via Halton sampling) from an original Multi-Class Support Vector Data Description algorithm. To evaluate the performance of the proposed method, we extracted a set of counterfactual explanations from three state-of-the-art datasets achieving satisfactory results that pave the way to a range of real-world applications. Impact Statement-When a system is analyzed by Artificial Intelligence, the inherent models are posed to the attention of domain experts, thus delegating further possible actions. Counterfactual explanations, on the other hand, directly suggest actuation on the system. Counterfactual control still remains under experts' supervision, but the system improves its level of autonomy. The long-term goal is to make the AI model aware of how to affect the environment properly (both in terms of performance and safety). Examples may include: manoeuvering of autonomous cars, clinical diagnosis, and finance. The proposed approach generalizes counterfactuals intelligibility and control to the multi-class case. The validation over practical scenarios (e.g., the FIFA dataset) corroborates both control precision and quality of counterfactual explanations, thus increasing the readiness level of the approach.

ACM Transactions on Intelligent Systems and Technology, 2019

Central to explanatory simulation models is their capability to not just show that but also why particular things happen. Explanation is closely related with the detection of causal relationships and is, in a simulation context, typically done by means of controlled experiments. However, for complex simulation models, conventional “blackbox” experiments may be too coarse-grained to cope with spurious relationships. We present an intervention-based causal analysis methodology that exploits the manipulability of computational models, and detects and circumvents spurious effects. The core of the methodology is a formal model that maps basic causal assumptions to causal observations and allows for the identification of combinations of assumptions that have a negative impact on observability. First, experiments indicate that the methodology can successfully deal with notoriously tricky situations involving asymmetric and symmetric overdetermination and detect fine-grained causal relation...

2021

Copyright held by the owner/author(s). CHI’21,, May 8–9, 2021, Virtual ACM 978-1-4503-6819-3/20/04. https://doi.org/10.1145/3334480.XXXXXXX Abstract Counterfactual explanations (CFEs) are an emerging technique under the umbrella of interpretability of machine learning (ML) models. They provide “what if” feedback of the form “if an input datapoint were x′ instead of x, then an ML model’s output would be y′ instead of y.” Counterfactual explainability for ML models has yet to see widespread adoption in industry. In this short paper, we posit reasons for this slow uptake. Leveraging recent work outlining desirable properties of CFEs and our experience running the ML wing of a model monitoring startup, we identify outstanding obstacles hindering CFE deployment in industry.