Efficient Explanations With Relevant Sets

Proceedings of the AAAI Conference on Artificial Intelligence

The complete reason behind a decision is a Boolean formula that characterizes why the decision was made. This recently introduced notion has a number of applications, which include generating explanations, detecting decision bias and evaluating counterfactual queries. Prime implicants of the complete reason are known as sufficient reasons for the decision and they correspond to what is known as PI explanations and abductive explanations. In this paper, we refer to the prime implicates of a complete reason as necessary reasons for the decision. We justify this terminology semantically and show that necessary reasons correspond to what is known as contrastive explanations. We also study the computation of complete reasons for multi-class decision trees and graphs with nominal and numeric features for which we derive efficient, closed-form complete reasons. We further investigate the computation of shortest necessary and sufficient reasons for a broad class of complete reasons, which i...

Proceedings of the AAAI Conference on Artificial Intelligence

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.

Data & Knowledge Engineering

Cornell University - arXiv, 2020

Recent work proposed the computation of so-called PIexplanations of Naive Bayes Classifiers (NBCs) [29]. PI-explanations are subset-minimal sets of feature-value pairs that are sufficient for the prediction, and have been computed with state-of-the-art exact algorithms that are worst-case exponential in time and space. In contrast, we show that the computation of one PI-explanation for an NBC can be achieved in log-linear time, and that the same result also applies to the more general class of linear classifiers. Furthermore, we show that the enumeration of PI-explanations can be obtained with polynomial delay. Experimental results demonstrate the performance gains of the new algorithms when compared with earlier work. The experimental results also investigate ways to measure the quality of heuristic explanations.

ArXiv, 2020

There is an increasing interest in and demand for interpretations and explanations of machine learning models and predictions in various application areas. In this paper, we consider data-driven models which are already developed, implemented and trained. Our goal is to interpret the models and explain and understand their predictions. Since the predictions made by data-driven models rely heavily on the data used for training, we believe explanations should convey information about how the training data affects the predictions. To do this, we propose a novel methodology which we call Shapley values for training data subset importance. The Shapley value concept originates from coalitional game theory, developed to fairly distribute the payout among a set of cooperating players. We extend this to subset importance, where a prediction is explained by treating the subsets of the training data as players in a game where the predictions are the payouts. We describe and illustrate how the ...

Semantic Web

Deep learning models have achieved impressive performance in various tasks, but they are usually opaque with regards to their inner complex operation, obfuscating the reasons for which they make decisions. This opacity raises ethical and legal concerns regarding the real-life use of such models, especially in critical domains such as in medicine, and has led to the emergence of the eXplainable Artificial Intelligence (XAI) field of research, which aims to make the operation of opaque AI systems more comprehensible to humans. The problem of explaining a black-box classifier is often approached by feeding it data and observing its behaviour. In this work, we feed the classifier with data that are part of a knowledge graph, and describe the behaviour with rules that are expressed in the terminology of the knowledge graph, that is understandable by humans. We first theoretically investigate the problem to provide guarantees for the extracted rules and then we investigate the relation of...

arXiv (Cornell University), 2022

Ensemble models (bagging and gradient-boosting) of relational decision trees have proved to be one of the most effective learning methods in the area of probabilistic logic models (PLMs). While effective, they lose one of the most important aspect of PLMs-interpretability. In this paper we consider the problem of compressing a large set of learned trees into a single explainable model. To this effect, we propose CoTE-Compression of Tree Ensembles-that produces a single small decision list as a compressed representation. CoTE first converts the trees to decision lists and then performs the combination and compression with the aid of the original training set. An experimental evaluation demonstrates the effectiveness of CoTE in several benchmark relational data sets.

Proceedings, 2020

In this paper we introduce aspBEEF, a tool for generating explanations for the outcome of an arbitrary machine learning classifier. This is done using Grover’s et al. framework known as Balanced English Explanations of Forecasts (BEEF) that generates explanations in terms of in terms of finite intervals over the values of the input features. Since the problem of obtaining an optimal BEEF explanation has been proved to be NP-complete, BEEF existing implementation computes an approximation. In this work we use instead an encoding into the Answer Set Programming paradigm, specialized in solving NP problems, to guarantee that the computed solutions are optimal.

Lecture Notes in Computer Science, 2021

The paper introduces a novel framework for extracting model-agnostic human interpretable rules to explain a classifier's output. The human interpretable rule is defined as an axis-aligned hyper-cuboid containing the instance for which the classification decision has to be explained. The proposed procedure finds the largest (high coverage) axis-aligned hyper-cuboid such that a high percentage of the instances in the hyper-cuboid have the same class label as the instance being explained (high precision). Novel approximations to the coverage and precision measures in terms of the parameters of the hyper-cuboid are defined. They are maximized using gradient-based optimizers. The quality of the approximations is rigorously analyzed theoretically and experimentally. Heuristics for simplifying the generated explanations for achieving better interpretability and a greedy selection algorithm that combines the local explanations for creating global explanations for the model covering a large part of the instance space are also proposed. The framework is model agnostic, can be applied to any arbitrary classifier, and all types of attributes (including continuous, ordered, and unordered discrete). The wide-scale applicability of the framework is validated on a variety of synthetic and real-world datasets from different domains (tabular, text, and image).

ArXiv, 2020

Many important questions about a model cannot be answered just by explaining how much each feature contributes to its output. To answer a broader set of questions, we generalize a popular, mathematically well-grounded explanation technique, Shapley Additive Explanations (SHAP). Our new method - Generalized Shapley Additive Explanations (G-SHAP) - produces many additional types of explanations, including: 1) General classification explanations; Why is this sample more likely to belong to one class rather than another? 2) Intergroup differences; Why do our model's predictions differ between groups of observations? 3) Model failure; Why does our model perform poorly on a given sample? We formally define these types of explanations and illustrate their practical use on real data.