Causal Explanations for Image Classifiers

Existing algorithms for explaining the output of image classifiers use different definitions of explanations and a variety of techniques to extract them. However, none of the existing tools use a principled approach based on formal definitions of causes and explanations for the explanation extraction. In this paper we present a novel black-box approach to computing explanations grounded in the theory of actual causality. We prove relevant theoretical results and present an algorithm for computing approximate explanations based on these definitions. We prove termination of our algorithm and discuss its complexity and the amount of approximation compared to the precise definition. We implemented the framework in a tool rex and we present experimental results and a comparison with state-of-the-art tools. We demonstrate that rex is the most efficient tool and produces the smallest explanations, in addition to outperforming other black-box tools on standard quality measures.

Test Where Decisions Matter: Importance-driven Testing for Deep Reinforcement Learning

In many Deep Reinforcement Learning (RL) problems, decisions in a trained policy vary in significance for the expected safety and performance of the policy. Since RL policies are very complex, testing efforts should concentrate on states in which the agent's decisions have the highest impact on the expected outcome. In this paper, we propose a novel model-based method to rigorously compute a ranking of state importance across the entire state space. We then focus our testing efforts on the highest-ranked states. In this paper, we focus on testing for safety. However, the proposed methods can be easily adapted to test for performance. In each iteration, our testing framework computes optimistic and pessimistic safety estimates. These estimates provide lower and upper bounds on the expected outcomes of the policy execution across all modeled states in the state space. Our approach divides the state space into safe and unsafe regions upon convergence, providing clear insights into the policy's weaknesses. Two important properties characterize our approach. (1) Optimal Test-Case Selection: At any time in the testing process, our approach evaluates the policy in the states that are most critical for safety. (2) Guaranteed Safety: Our approach can provide formal verification guarantees over the entire state space by sampling only a fraction of the policy. Any safety properties assured by the pessimistic estimate are formally proven to hold for the policy. We provide a detailed evaluation of our framework on several examples, showing that our method discovers unsafe policy behavior with low testing effort.

Real-Time Incremental Explanations for Object Detectors

Existing black box explainability tools for object detectors rely on multiple calls to the model, which prevents them from computing explanations in real time. In this paper we introduce IncX, an algorithm for real-time incremental approximations of explanations, based on linear transformations of saliency maps. We implement IncX on top of D-RISE, a state-of-the-art black-box explainability tool for object detectors. We show that IncX's explanations are comparable in quality to those of D-RISE, with insertion curves being within 8%, and are computed two orders of magnitude faster that D-RISE's explanations.

Counterfactual Influence in Markov Decision Processes

Our work addresses a fundamental problem in the context of counterfactual inference for Markov Decision Processes (MDPs). Given an MDP path $\tau$, this kind of inference allows us to derive counterfactual paths $\tau'$ describing what-if versions of $\tau$ obtained under different action sequences than those observed in $\tau$. However, as the counterfactual states and actions deviate from the observed ones over time, the observation $\tau$ may no longer influence the counterfactual world, meaning that the analysis is no longer tailored to the individual observation, resulting in interventional outcomes rather than counterfactual ones. Even though this issue specifically affects the popular Gumbel-max structural causal model used for MDP counterfactuals, it has remained overlooked until now. In this work, we introduce a formal characterisation of influence based on comparing counterfactual and interventional distributions. We devise an algorithm to construct counterfactual models that automatically satisfy influence constraints. Leveraging such models, we derive counterfactual policies that are not just optimal for a given reward structure but also remain tailored to the observed path. Even though there is an unavoidable trade-off between policy optimality and strength of influence constraints, our experiments demonstrate that it is possible to derive (near-)optimal policies while remaining under the influence of the observation.

Explaining Image Classifiers

We focus on explaining image classifiers, taking the work of Mothilal et al. [2021] (MMTS) as our point of departure. We observe that, although MMTS claim to be using the definition of explanation proposed by Halpern [2016], they do not quite do so. Roughly speaking, Halpern's definition has a necessity clause and a sufficiency clause. MMTS replace the necessity clause by a requirement that, as we show, implies it. Halpern's definition also allows agents to restrict the set of options considered. While these difference may seem minor, as we show, they can have a nontrivial impact on explanations. We also show that, essentially without change, Halpern's definition can handle two issues that have proved difficult for other approaches: explanations of absence (when, for example, an image classifier for tumors outputs "no tumor") and explanations of rare events (such as tumors).

MRxaI: Black-Box Explainability for Image Classifiers in a Medical Setting

Existing tools for explaining the output of image classifiers can be divided into white-box, which rely on access to the model internals, and black-box, agnostic to the model. As the usage of AI in the medical domain grows, so too does the usage of explainability tools. Existing work on medical image explanations focuses on white-box tools, such as gradcam. However, there are clear advantages to switching to a black-box tool, including the ability to use it with any classifier and the wide selection of black-box tools available. On standard images, black-box tools are as precise as white-box. In this paper we compare the performance of several black-box methods against gradcam on a brain cancer MRI dataset. We demonstrate that most black-box tools are not suitable for explaining medical image classifications and present a detailed analysis of the reasons for their shortcomings. We also show that one black-box tool, a causal explainability-based rex, performs as well as \gradcam.

You Only Explain Once

In this paper, we propose a new black-box explainability algorithm and tool, YO-ReX, for efficient explanation of the outputs of object detectors. The new algorithm computes explanations for all objects detected in the image simultaneously. Hence, compared to the baseline, the new algorithm reduces the number of queries by a factor of 10X for the case of ten detected objects. The speedup increases further with with the number of objects. Our experimental results demonstrate that YO-ReX can explain the outputs of YOLO with a negligible overhead over the running time of YOLO. We also demonstrate similar results for explaining SSD and Faster R-CNN. The speedup is achieved by avoiding backtracking by combining aggressive pruning with a causal analysis.

Clustered Policy Decision Ranking

Policies trained via reinforcement learning (RL) are often very complex even for simple tasks. In an episode with n time steps, a policy will make n decisions on actions to take, many of which may appear non-intuitive to the observer. Moreover, it is not clear which of these decisions directly contribute towards achieving the reward and how significant their contribution is. Given a trained policy, we propose a black-box method based on statistical covariance estimation that clusters the states of the environment and ranks each cluster according to the importance of decisions made in its states. We compare our measure against a previous statistical fault localization based ranking procedure.

Multiple Different Explanations for Image Classifiers

Existing explanation tools for image classifiers usually give only one single explanation for an image. For many images, however, both humans and image classifiers accept more than one explanation for the image label. Thus, restricting the number of explanations to just one severely limits the insight into the behavior of the classifier. In this paper, we describe an algorithm and a tool, REX, for computing multiple explanations of the output of a black-box image classifier for a given image. Our algorithm uses a principled approach based on causal theory. We analyse its theoretical complexity and provide experimental results showing that REX finds multiple explanations on 7 times more images than the previous work on the ImageNet-mini benchmark.

Equality of Effort via Algorithmic Recourse

This paper proposes a method for measuring fairness through equality of effort by applying algorithmic recourse through minimal interventions. Equality of effort is a property that can be quantified at both the individual and the group level. It answers the counterfactual question: what is the minimal cost for a protected individual or the average minimal cost for a protected group of individuals to reverse the outcome computed by an automated system? Algorithmic recourse increases the flexibility and applicability of the notion of equal effort: it overcomes its previous limitations by reconciling multiple treatment variables, introducing feasibility and plausibility constraints, and integrating the actual relative costs of interventions. We extend the existing definition of equality of effort and present an algorithm for its assessment via algorithmic recourse. We validate our approach both on synthetic data and on the German credit dataset.