Explainable Artificial Intelligence (XAI) is critical for attaining trust in the operation of AI systems. A key question of an AI system is ``why was this decision made this way''. Formal approaches to XAI use a formal model of the AI system to identify abductive explanations. While abductive explanations may be applicable to a large number of inputs sharing the same concrete values, more general explanations may be preferred for numeric inputs. So-called inflated abductive explanations give intervals for each feature ensuring that any input whose values fall withing these intervals is still guaranteed to make the same prediction. Inflated explanations cover a larger portion of the input space, and hence are deemed more general explanations. But there can be many (inflated) abductive explanations for an instance. Which is the best? In this paper, we show how to find a most general abductive explanation for an AI decision. This explanation covers as much of the input space as possible, while still being a correct formal explanation of the model's behaviour. Given that we only want to give a human one explanation for a decision, the most general explanation gives us the explanation with the broadest applicability, and hence the one most likely to seem sensible.

The recent criticisms of the robustness of post hoc model approximation explanation methods (like LIME and SHAP) have led to the rise of model-precise abductive explanations. For each data point, abductive explanations provide a minimal subset of features that are sufficient to generate the outcome. While theoretically sound and rigorous, abductive explanations suffer from a major issue --- there can be several valid abductive explanations for the same data point. In such cases, providing a single abductive explanation can be insufficient; on the other hand, providing all valid abductive explanations can be incomprehensible due to their size. In this work, we solve this issue by aggregating the many possible abductive explanations into feature importance scores. We propose three aggregation methods: two based on power indices from cooperative game theory and a third based on a well-known measure of causal strength. We characterize these three methods axiomatically, showing that each of them uniquely satisfies a set of desirable properties. We also evaluate them on multiple datasets and show that these explanations are robust to the attacks that fool SHAP and LIME.

In the quest for Explainable Artificial Intelligence (XAI) one of the questions that frequently arises given a decision made by an AI system is, ``why was the decision made in this way?'' Formal approaches to explainability build a formal model of the AI system and use this to reason about the properties of the system. Given a set of feature values for an instance to be explained, and a resulting decision, a formal abductive explanation is a set of features, such that if they take the given value will always lead to the same decision. This explanation is useful, it shows that only some features were used in making the final decision. But it is narrow, it only shows that if the selected features take their given values the decision is unchanged. It is possible that some features may change values and still lead to the same decision. In this paper we formally define inflated explanations which is a set of features, and for each feature a set of values (always including the value of the instance being explained), such that the decision will remain unchanged, for any of the values allowed for any of the features in the (inflated) abductive explanation. Inflated formal explanations are more informative than common abductive explanations since e.g. they allow us to see if the exact value of a feature is important, or it could be any nearby value. Overall they allow us to better understand the role of each feature in the decision. We show that we can compute inflated explanations for not that much greater cost than abductive explanations, and that we can extend duality results for abductive explanations also to inflated explanations.

Explainable Artificial Intelligence (XAI) is widely regarded as a cornerstone of trustworthy AI. Unfortunately, most work on XAI offers no guarantees of rigor. In high-stakes domains, e.g. uses of AI that impact humans, the lack of rigor of explanations can have disastrous consequences. Formal abductive explanations offer crucial guarantees of rigor and so are of interest in high-stakes uses of machine learning (ML). One drawback of abductive explanations is explanation size, justified by the cognitive limits of human decision-makers. Probabilistic abductive explanations (PAXps) address this limitation, but their theoretical and practical complexity makes their exact computation most often unrealistic. This paper proposes novel efficient algorithms for the computation of locally-minimal PAXps, which offer high-quality approximations of PXAps in practice. The experimental results demonstrate the practical efficiency of the proposed solutions.

The uses of machine learning (ML) have snowballed in recent years. In many cases, ML models are highly complex, and their operation is beyond the understanding of human decision-makers. Nevertheless, some uses of ML models involve high-stakes and safety-critical applications. Explainable artificial intelligence (XAI) aims to help human decision-makers in understanding the operation of such complex ML models, thus eliciting trust in their operation. Unfortunately, the majority of past XAI work is based on informal approaches, that offer no guarantees of rigor. Unsurprisingly, there exists comprehensive experimental and theoretical evidence confirming that informal methods of XAI can provide human-decision makers with erroneous information. Logic-based XAI represents a rigorous approach to explainability; it is model-based and offers the strongest guarantees of rigor of computed explanations. However, a well-known drawback of logic-based XAI is the complexity of logic reasoning, especially for highly complex ML models. Recent work proposed distance-restricted explanations, i.e. explanations that are rigorous provided the distance to a given input is small enough. Distance-restricted explainability is tightly related with adversarial robustness, and it has been shown to scale for moderately complex ML models, but the number of inputs still represents a key limiting factor. This paper investigates novel algorithms for scaling up the performance of logic-based explainers when computing and enumerating ML model explanations with a large number of inputs.

Trustable explanations of machine learning (ML) models are vital in high-risk uses of artificial intelligence (AI). Apart from the computation of trustable explanations, a number of explainability queries have been identified and studied in recent work. Some of these queries involve solving quantification problems, either in propositional or in more expressive logics. This paper investigates one of these quantification problems, namely the feature relevancy problem (FRP), i.e.\ to decide whether a (possibly sensitive) feature can occur in some explanation of a prediction. In contrast with earlier work, that studied FRP for specific classifiers, this paper proposes a novel algorithm for the \fprob quantification problem which is applicable to any ML classifier that meets minor requirements. Furthermore, the paper shows that the novel algorithm is efficient in practice. The experimental results, obtained using random forests (RFs) induced from well-known publicly available datasets, demonstrate that the proposed solution outperforms existing state-of-the-art solvers for Quantified Boolean Formulas (QBF) by orders of magnitude. Finally, the paper also identifies a novel family of formulas that are challenging for currently state-of-the-art QBF solvers.

Claims about the interpretability of decision trees can be traced back to the origins of machine learning (ML). Indeed, given some input consistent with a decision tree's path, the explanation for the resulting prediction consists of the features in that path. Moreover, a growing number of works propose the use of decision trees, and of other so-called interpretable models, as a possible solution for deploying ML models in high-risk applications. This paper overviews recent theoretical and practical results which demonstrate that for most decision trees, tree paths exhibit so-called explanation redundancy, in that logically sound explanations can often be significantly more succinct than what the features in the path dictates. More importantly, such decision tree explanations can be computed in polynomial-time, and so can be produced with essentially no effort other than traversing the decision tree. The experimental results, obtained on a large range of publicly available decision trees, support the paper's claims.

Compilation into propositional languages finds a growing number of practical uses, including in constraint programming, diagnosis and machine learning (ML), among others. One concrete example is the use of propositional languages as classifiers, and one natural question is how to explain the predictions made. This paper shows that for classifiers represented with some of the best-known propositional languages, different kinds of explanations can be computed in polynomial time. These languages include deterministic decomposable negation normal form (d-DNNF), and so any propositional language that is strictly less succinct than d-DNNF. Furthermore, the paper describes optimizations, specific to Sentential Decision Diagrams (SDDs), which are shown to yield more efficient algorithms in practice.

Tree ensembles (TEs) denote a prevalent machine learning model that do not offer guarantees of interpretability, that represent a challenge from the perspective of explainable artificial intelligence. Besides model agnostic approaches, recent work proposed to explain TEs with formally-defined explanations, which are computed with oracles for propositional satisfiability (SAT) and satisfiability modulo theories. The computation of explanations for TEs involves linear constraints to express the prediction. In practice, this deteriorates scalability of the underlying reasoners. Motivated by the inherent propositional nature of TEs, this paper proposes to circumvent the need for linear constraints and instead employ an optimization engine for pure propositional logic to efficiently handle the prediction. Concretely, the paper proposes to use a MaxSAT solver and exploit the objective function to determine a winning class. This is achieved by devising a propositional encoding for computing explanations of TEs. Furthermore, the paper proposes additional heuristics to improve the underlying MaxSAT solving procedure. Experimental results obtained on a wide range of publicly available datasets demonstrate that the proposed MaxSAT-based approach is either on par or outperforms the existing reasoning-based explainers, thus representing a robust and efficient alternative for computing formal explanations for TEs.

Decision trees (DTs) epitomize the ideal of interpretability of machine learning (ML) models. The interpretability of decision trees motivates explainability approaches by so-called intrinsic interpretability, and it is at the core of recent proposals for applying interpretable ML models in high-risk applications. The belief in DT interpretability is justified by the fact that explanations for DT predictions are generally expected to be succinct. Indeed, in the case of DTs, explanations correspond to DT paths. Since decision trees are ideally shallow, and so paths contain far fewer features than the total number of features, explanations in DTs are expected to be succinct, and hence interpretable. This paper offers both theoretical and experimental arguments demonstrating that, as long as interpretability of decision trees equates with succinctness of explanations, then decision trees ought not be deemed interpretable. The paper introduces logically rigorous path explanations and path explanation redundancy, and proves that there exist functions for which decision trees must exhibit paths with explanation redundancy that is arbitrarily larger than the actual path explanation. The paper also proves that only a very restricted class of functions can be represented with DTs that exhibit no explanation redundancy. In addition, the paper includes experimental results substantiating that path explanation redundancy is observed ubiquitously in decision trees, including those obtained using different tree learning algorithms, but also in a wide range of publicly available decision trees. The paper also proposes polynomial-time algorithms for eliminating path explanation redundancy, which in practice require negligible time to compute. Thus, these algorithms serve to indirectly attain irreducible, and so succinct, explanations for decision trees. Furthermore, the paper includes novel results related with duality and enumeration of explanations, based on using SAT solvers as witness-producing NP-oracles.

Random Forest (RFs) are among the most widely used Machine Learning (ML) classifiers. Even though RFs are not interpretable, there are no dedicated non-heuristic approaches for computing explanations of RFs. Moreover, there is recent work on polynomial algorithms for explaining ML models, including naive Bayes classifiers. Hence, one question is whether finding explanations of RFs can be solved in polynomial time. This paper answers this question negatively, by proving that computing one PI-explanation of an RF is D^P-hard. Furthermore, the paper proposes a propositional encoding for computing explanations of RFs, thus enabling finding PI-explanations with a SAT solver. This contrasts with earlier work on explaining boosted trees (BTs) and neural networks (NNs), which requires encodings based on SMT/MILP. Experimental results, obtained on a wide range of publicly available datasets, demonstrate that the proposed SAT-based approach scales to RFs of sizes common in practical applications. Perhaps more importantly, the experimental results demonstrate that, for the vast majority of examples considered, the SAT-based approach proposed in this paper significantly outperforms existing heuristic approaches.

Recent work has shown that not only decision trees (DTs) may not be interpretable but also proposed a polynomial-time algorithm for computing one PI-explanation of a DT. This paper shows that for a wide range of classifiers, globally referred to as decision graphs, and which include decision trees and binary decision diagrams, but also their multi-valued variants, there exist polynomial-time algorithms for computing one PI-explanation. In addition, the paper also proposes a polynomial-time algorithm for computing one contrastive explanation. These novel algorithms build on explanation graphs (XpG's). XpG's denote a graph representation that enables both theoretical and practically efficient computation of explanations for decision graphs. Furthermore, the paper proposes a practically efficient solution for the enumeration of explanations, and studies the complexity of deciding whether a given feature is included in some explanation. For the concrete case of decision trees, the paper shows that the set of all contrastive explanations can be enumerated in polynomial time. Finally, the experimental results validate the practical applicability of the algorithms proposed in the paper on a wide range of publicly available benchmarks.

International audience

While traditional data mining techniques have been used extensively for finding patterns in databases, they are not always suitable for incorporating user-specified constraints. To overcome this issue, CP and SAT based frameworks for modeling and solving pattern mining tasks have gained a considerable audience in recent years. However, a bottleneck for all these CP and SAT-based approaches is the encoding size which makes these algorithms inefficient for large databases. This paper introduces a practical SAT-based approach to discover efficiently (minimal non-redundant) association rules. First, we present a decomposition-based paradigm that splits the original transaction database into smaller and independent subsets. Then, we show that without producing too large formulas, our decomposition method allows independent mining evaluation on a multi-core machine, improving performance. Finally, an experimental evaluation shows that our method is fast and scale well compared with the existing CP approach even in the sequential case, while significantly reducing the gap with the best state-of-the-art specialized algorithm.

The aim of this paper is to highlight two important issues related to the annotation and querying of Intangible Cultural Heritage video datasets. First, we focus on ontology completion by annotating dance videos. In order to build video training sets and to enrich the proposed ontology, manual video annotation is performed based on background knowledge formalized in an ontology, representing a semantics of a traditional dance. The paper provides a case study on Malaysian Zapin dances. Second, we address the question of how can end-users efficiently query the datasets of annotated videos that are built.

The enumeration of all Maximal Satisfiable Subsets (MSSes) or all Minimal Correction Subsets (MCSes) of an unsatisfiable CNF Boolean formula is a useful and sometimes necessary step for solving a variety of important A.I. issues. Although the number of different MCSes of a CNF Boolean formula is exponential in the worst case, it remains low in many practical situations; this makes the tentative enumeration possibly successful in these latter cases. In the paper, a technique is introduced that boosts the currently most efficient practical approaches to enumerate MCSes. It implements a model rotation paradigm that allows the set of MCSes to be computed in an heuristically efficient way.

The search for one consensus that reconciles several conflicting agents or information sources is a key paradigm in both everyday-life and A.I. In this paper, we improve a recent approach to consensus-finding that has been developed in the Boolean logic framework. In that recent approach consensuses are defined as being some specific non-conflicting fragments of the set-theoretic union of all the information to be mitigated. In order to be a consensus, such a fragment must be conflict-free with every information source, taken individually. That approach has also introduced a generic method to compute one consensus that obeys various possible maximality criteria, including maximality with respect to set-theoretical inclusion, i.e., max⊆ consensuses. As max⊆ consensuses are of clear interest, the focus is on their computation in this paper. We develop an algorithm for computing one max⊆ consensus that is experimentally more time and space efficient than the computational method proposed in the initial study.

The efficient extraction of one maximal information subset that does not conflict with multiple contxts or additional information sources is a key basic issue in many A.I. domains, especially when these contexts or sources can be mutually conflicting. In this paper, this question is addressed from a computational point of view in clausal Boolean logic. A new approach is introduced that experimentally outperforms the currently most efficient technique.