Counterfactual explanation is one branch of interpretable machine learning that produces a perturbation sample to change the model's original decision. The generated samples can act as a recommendation for end-users to achieve their desired outputs. Most of the current counterfactual explanation approaches are the gradient-based method, which can only optimize the differentiable loss functions with continuous variables. Accordingly, the gradient-free methods are proposed to handle the categorical variables, which however have several major limitations: 1) causal relationships among features are typically ignored when generating the counterfactuals, possibly resulting in impractical guidelines for decision-makers; 2) the counterfactual explanation algorithm requires a great deal of effort into parameter tuning for dertermining the optimal weight for each loss functions which must be conducted repeatedly for different datasets and settings. In this work, to address the above limitations, we propose a prototype-based counterfactual explanation framework (ProCE). ProCE is capable of preserving the causal relationship underlying the features of the counterfactual data. In addition, we design a novel gradient-free optimization based on the multi-objective genetic algorithm that generates the counterfactual explanations for the mixed-type of continuous and categorical features. Numerical experiments demonstrate that our method compares favorably with state-of-the-art methods and therefore is applicable to existing prediction models. All the source codes and data are available at \url{https://github.com/tridungduong16/multiobj-scm-cf}.
Feature attribution methods are exceedingly popular in interpretable machine learning. They aim to compute the attribution of each input feature to represent its importance, but there is no consensus on the definition of "attribution", leading to many competing methods with little systematic evaluation. The lack of attribution ground truth further complicates evaluation, which has to rely on proxy metrics. To address this, we propose a dataset modification procedure such that models trained on the new dataset have ground truth attribution available. We evaluate three methods: saliency maps, rationales, and attention. We identify their deficiencies and add a new perspective to the growing body of evidence questioning their correctness and reliability in the wild. Our evaluation approach is model-agnostic and can be used to assess future feature attribution method proposals as well. Code is available at https://github.com/YilunZhou/feature-attribution-evaluation.
We take inspiration from the study of human explanation to inform the design and evaluation of interpretability methods in machine learning. First, we survey the literature on human explanation in philosophy, cognitive science, and the social sciences, and propose a list of design principles for machine-generated explanations that are meaningful to humans. Using the concept of weight of evidence from information theory, we develop a method for generating explanations that adhere to these principles. We show that this method can be adapted to handle high-dimensional, multi-class settings, yielding a flexible framework for generating explanations. We demonstrate that these explanations can be estimated accurately from finite samples and are robust to small perturbations of the inputs. We also evaluate our method through a qualitative user study with machine learning practitioners, where we observe that the resulting explanations are usable despite some participants struggling with background concepts like prior class probabilities. Finally, we conclude by surfacing~design~implications for interpretability tools in general.
Feature attribution is often loosely presented as the process of selecting a subset of relevant features as a rationale of a prediction. Task-dependent by nature, precise definitions of "relevance" encountered in the literature are however not always consistent. This lack of clarity stems from the fact that we usually do not have access to any notion of ground-truth attribution and from a more general debate on what good interpretations are. In this paper we propose to formalise feature selection/attribution based on the concept of relaxed functional dependence. In particular, we extend our notions to the instance-wise setting and derive necessary properties for candidate selection solutions, while leaving room for task-dependence. By computing ground-truth attributions on synthetic datasets, we evaluate many state-of-the-art attribution methods and show that, even when optimised, some fail to verify the proposed properties and provide wrong solutions.
Artificial Intelligence (AI) has a tremendous impact on the unexpected growth of technology in almost every aspect. AI-powered systems are monitoring and deciding about sensitive economic and societal issues. The future is towards automation, and it must not be prevented. However, this is a conflicting viewpoint for a lot of people, due to the fear of uncontrollable AI systems. This concern could be reasonable if it was originating from considerations associated with social issues, like gender-biased, or obscure decision-making systems. Explainable AI (XAI) is recently treated as a huge step towards reliable systems, enhancing the trust of people to AI. Interpretable machine learning (IML), a subfield of XAI, is also an urgent topic of research. This paper presents a small but significant contribution to the IML community, focusing on a local-based, neural-specific interpretation process applied to textual and time-series data. The proposed methodology introduces new approaches to the presentation of feature importance based interpretations, as well as the production of counterfactual words on textual datasets. Eventually, an improved evaluation metric is introduced for the assessment of interpretation techniques, which supports an extensive set of qualitative and quantitative experiments.
One of the key elements of explanatory analysis of a predictive model is to assess the importance of individual variables. Rapid development of the area of predictive model exploration (also called explainable artificial intelligence or interpretable machine learning) has led to the popularization of methods for local (instance level) and global (dataset level) methods, such as Permutational Variable Importance, Shapley Values (SHAP), Local Interpretable Model Explanations (LIME), Break Down and so on. However, these methods do not use information about the correlation between features which significantly reduce the explainability of the model behaviour. In this work, we propose new methods to support model analysis by exploiting the information about the correlation between variables. The dataset level aspect importance measure is inspired by the block permutations procedure, while the instance level aspect importance measure is inspired by the LIME method. We show how to analyze groups of variables (aspects) both when they are proposed by the user and when they should be determined automatically based on the hierarchical structure of correlations between variables. Additionally, we present the new type of model visualisation, triplot, which exploits a hierarchical structure of variable grouping to produce a high information density model visualisation. This visualisation provides a consistent illustration for either local or global model and data exploration. We also show an example of real-world data with 5k instances and 37 features in which a significant correlation between variables affects the interpretation of the effect of variable importance. The proposed method is, to our knowledge, the first to allow direct use of the correlation between variables in exploratory model analysis.
Disentanglement, a critical concern in interpretable machine learning, has also garnered significant attention from the computer vision community. Many existing GAN-based class disentanglement (unsupervised) approaches, such as InfoGAN and its variants, primarily aim to maximize the mutual information (MI) between the generated image and its latent codes. However, this focus may lead to a tendency for the network to generate highly similar images when presented with the same latent class factor, potentially resulting in mode collapse or mode dropping. To alleviate this problem, we propose \texttt{CoDeGAN} (Contrastive Disentanglement for Generative Adversarial Networks), where we relax similarity constraints for disentanglement from the image domain to the feature domain. This modification not only enhances the stability of GAN training but also improves their disentangling capabilities. Moreover, we integrate self-supervised pre-training into CoDeGAN to learn semantic representations, significantly facilitating unsupervised disentanglement. Extensive experimental results demonstrate the superiority of our method over state-of-the-art approaches across multiple benchmarks. The code is available at https://github.com/learninginvision/CoDeGAN.
While the need for interpretable machine learning has been established, many common approaches are slow, lack fidelity, or hard to evaluate. Amortized explanation methods reduce the cost of providing interpretations by learning a global selector model that returns feature importances for a single instance of data. The selector model is trained to optimize the fidelity of the interpretations, as evaluated by a predictor model for the target. Popular methods learn the selector and predictor model in concert, which we show allows predictions to be encoded within interpretations. We introduce EVAL-X as a method to quantitatively evaluate interpretations and REAL-X as an amortized explanation method, which learn a predictor model that approximates the true data generating distribution given any subset of the input. We show EVAL-X can detect when predictions are encoded in interpretations and show the advantages of REAL-X through quantitative and radiologist evaluation.
Deep learning has been the engine powering many successes of data science. However, the deep neural network (DNN), as the basic model of deep learning, is often excessively over-parameterized, causing many difficulties in training, prediction and interpretation. We propose a frequentist-like method for learning sparse DNNs and justify its consistency under the Bayesian framework: the proposed method could learn a sparse DNN with at most $O(n/\log(n))$ connections and nice theoretical guarantees such as posterior consistency, variable selection consistency and asymptotically optimal generalization bounds. In particular, we establish posterior consistency for the sparse DNN with a mixture Gaussian prior, show that the structure of the sparse DNN can be consistently determined using a Laplace approximation-based marginal posterior inclusion probability approach, and use Bayesian evidence to elicit sparse DNNs learned by an optimization method such as stochastic gradient descent in multiple runs with different initializations. The proposed method is computationally more efficient than standard Bayesian methods for large-scale sparse DNNs. The numerical results indicate that the proposed method can perform very well for large-scale network compression and high-dimensional nonlinear variable selection, both advancing interpretable machine learning.
High-resolution mapping of cells and tissue structures provides a foundation for developing interpretable machine-learning models for computational pathology. Deep learning algorithms can provide accurate mappings given large numbers of labeled instances for training and validation. Generating adequate volume of quality labels has emerged as a critical barrier in computational pathology given the time and effort required from pathologists. In this paper we describe an approach for engaging crowds of medical students and pathologists that was used to produce a dataset of over 220,000 annotations of cell nuclei in breast cancers. We show how suggested annotations generated by a weak algorithm can improve the accuracy of annotations generated by non-experts and can yield useful data for training segmentation algorithms without laborious manual tracing. We systematically examine interrater agreement and describe modifications to the MaskRCNN model to improve cell mapping. We also describe a technique we call Decision Tree Approximation of Learned Embeddings (DTALE) that leverages nucleus segmentations and morphologic features to improve the transparency of nucleus classification models. The annotation data produced in this study are freely available for algorithm development and benchmarking at: https://sites.google.com/view/nucls.
In this paper we introduce a tool called Principal Image Sections Mapping - PRISM, dedicated for PyTorch, but can be easily ported to other deep learning frameworks. Presented software relies on Principal Component Analysis to visualize the most significant features recognized by a given Convolutional Neural Network. Moreover, it allows to display comparative set features between images processed in the same batch, therefore PRISM can be a method well synerging with technique Explanation by Example.
The most important goal of customer services is to keep the customer satisfied. However, service resources are always limited and must be prioritized. Therefore, it is important to identify customers who potentially become unsatisfied and might lead to escalations. Today this prioritization of customers is often done manually. Data science on IoT data (esp. log data) for machine health monitoring, as well as analytics on enterprise data for customer relationship management (CRM) have mainly been researched and applied independently. In this paper, we present a framework for a data-driven decision support system which combines IoT and enterprise data to model customer sentiment. Such decision support systems can help to prioritize customers and service resources to effectively troubleshoot problems or even avoid them. The framework is applied in a real-world case study with a major medical device manufacturer. This includes a fully automated and interpretable machine learning pipeline designed to meet the requirements defined with domain experts and end users. The overall framework is currently deployed, learns and evaluates predictive models from terabytes of IoT and enterprise data to actively monitor the customer sentiment for a fleet of thousands of high-end medical devices. Furthermore, we provide an anonymized industrial benchmark dataset for the research community.
Genetic pathways usually encode molecular mechanisms that can inform targeted interventions. It is often challenging for existing machine learning approaches to jointly model genetic pathways (higher-order features) and variants (atomic features), and present to clinicians interpretable models. In order to build more accurate and better interpretable machine learning models for genetic medicine, we introduce Pathway Augmented Nonnegative Tensor factorization for HighER-order feature learning (PANTHER). PANTHER selects informative genetic pathways that directly encode molecular mechanisms. We apply genetically motivated constrained tensor factorization to group pathways in a way that reflects molecular mechanism interactions. We then train a softmax classifier for disease types using the identified pathway groups. We evaluated PANTHER against multiple state-of-the-art constrained tensor/matrix factorization models, as well as group guided and Bayesian hierarchical models. PANTHER outperforms all state-of-the-art comparison models significantly (p<0.05). Our experiments on large scale Next Generation Sequencing (NGS) and whole-genome genotyping datasets also demonstrated wide applicability of PANTHER. We performed feature analysis in predicting disease types, which suggested insights and benefits of the identified pathway groups.
In the time-series analysis, the time series motifs and the order patterns in time series can reveal general temporal patterns and dynamic features. Triadic Motif Field (TMF) is a simple and effective time-series image encoding method based on triadic time series motifs. Electrocardiography (ECG) signals are time-series data widely used to diagnose various cardiac anomalies. The TMF images contain the features characterizing the normal and Atrial Fibrillation (AF) ECG signals. Considering the quasi-periodic characteristics of ECG signals, the dynamic features can be extracted from the TMF images with the transfer learning pre-trained convolutional neural network (CNN) models. With the extracted features, the simple classifiers, such as the Multi-Layer Perceptron (MLP), the logistic regression, and the random forest, can be applied for accurate anomaly detection. With the test dataset of the PhysioNet Challenge 2017 database, the TMF classification model with the VGG16 transfer learning model and MLP classifier demonstrates the best performance with the 95.50% ROC-AUC and 88.43% F1 score in the AF classification. Besides, the TMF classification model can identify AF patients in the test dataset with high precision. The feature vectors extracted from the TMF images show clear patient-wise clustering with the t-distributed Stochastic Neighbor Embedding technique. Above all, the TMF classification model has very good clinical interpretability. The patterns revealed by symmetrized Gradient-weighted Class Activation Mapping have a clear clinical interpretation at the beat and rhythm levels.
As machine learning and algorithmic decision making systems are increasingly being leveraged in high-stakes human-in-the-loop settings, there is a pressing need to understand the rationale of their predictions. Researchers have responded to this need with explainable AI (XAI), but often proclaim interpretability axiomatically without evaluation. When these systems are evaluated, they are often tested through offline simulations with proxy metrics of interpretability (such as model complexity). We empirically evaluate the veracity of three common interpretability assumptions through a large scale human-subjects experiment with a simple "placebo explanation" control. We find that feature attribution explanations provide marginal utility in our task for a human decision maker and in certain cases result in worse decisions due to cognitive and contextual confounders. This result challenges the assumed universal benefit of applying these methods and we hope this work will underscore the importance of human evaluation in XAI research. Supplemental materials -- including anonymized data from the experiment, code to replicate the study, an interactive demo of the experiment, and the models used in the analysis -- can be found at: https://doi.pizza/challenging-xai.
Prototype-based methods use interpretable representations to address the black-box nature of deep learning models, in contrast to post-hoc explanation methods that only approximate such models. We propose the Neural Prototype Tree (ProtoTree), an intrinsically interpretable deep learning method for fine-grained image recognition. ProtoTree combines prototype learning with decision trees, and thus results in a globally interpretable model by design. Additionally, ProtoTree can locally explain a single prediction by outlining a decision path through the tree. Each node in our binary tree contains a trainable prototypical part. The presence or absence of this learned prototype in an image determines the routing through a node. Decision making is therefore similar to human reasoning: Does the bird have a red throat? And an elongated beak? Then it's a hummingbird! We tune the accuracy-interpretability trade-off using ensemble methods, pruning and binarizing. We apply pruning without sacrificing accuracy, resulting in a small tree with only 8 learned prototypes along a path to classify a bird from 200 species. An ensemble of 5 ProtoTrees achieves competitive accuracy on the CUB-200- 2011 and Stanford Cars data sets. Code is available at https://github.com/M-Nauta/ProtoTree
COVID-19 is not a universal killer. We study the spread of COVID-19 at the county level for the United States up until the 15$^{th}$ of August, 2020. We show that the prevalence of the disease and the death rate are correlated with the local socio-economic conditions often going beyond local population density distributions, especially in rural areas. We correlate the COVID-19 prevalence and death rate with data from the US Census Bureau and point out how the spreading patterns of the disease show asymmetries in urban and rural areas separately and are preferentially affecting the counties where a large fraction of the population is non-white. Our findings can be used for more targeted policy building and deployment of resources for future occurrence of a pandemic due to SARS-CoV-2. Our methodology, based on interpretable machine learning and game theory, can be extended to study the spread of other diseases.
The definition of Linear Symmetry-Based Disentanglement (LSBD) formalizes the notion of linearly disentangled representations, but there is currently no metric to quantify LSBD. Such a metric is crucial to evaluate LSBD methods and to compare to previous understandings of disentanglement. We propose $\mathcal{D}_\mathrm{LSBD}$, a mathematically sound metric to quantify LSBD, and provide a practical implementation for $\mathrm{SO}(2)$ groups. Furthermore, from this metric we derive LSBD-VAE, a semi-supervised method to learn LSBD representations. We demonstrate the utility of our metric by showing that (1) common VAE-based disentanglement methods don't learn LSBD representations, (2) LSBD-VAE as well as other recent methods can learn LSBD representations, needing only limited supervision on transformations, and (3) various desirable properties expressed by existing disentanglement metrics are also achieved by LSBD representations.
A method for the local and global interpretation of a black-box model on the basis of the well-known generalized additive models is proposed. It can be viewed as an extension or a modification of the algorithm using the neural additive model. The method is based on using an ensemble of gradient boosting machines (GBMs) such that each GBM is learned on a single feature and produces a shape function of the feature. The ensemble is composed as a weighted sum of separate GBMs resulting a weighted sum of shape functions which form the generalized additive model. GBMs are built in parallel using randomized decision trees of depth 1, which provide a very simple architecture. Weights of GBMs as well as features are computed in each iteration of boosting by using the Lasso method and then updated by means of a specific smoothing procedure. In contrast to the neural additive model, the method provides weights of features in the explicit form, and it is simply trained. A lot of numerical experiments with an algorithm implementing the proposed method on synthetic and real datasets demonstrate its efficiency and properties for local and global interpretation.
Explainable AI is an emerging field providing solutions for acquiring insights into automated systems' rationale. It has been put on the AI map by suggesting ways to tackle key ethical and societal issues. Existing explanation techniques are often not comprehensible to the end user. Lack of evaluation and selection criteria also makes it difficult for the end user to choose the most suitable technique. In this study, we combine logic-based argumentation with Interpretable Machine Learning, introducing a preliminary meta-explanation methodology that identifies the truthful parts of feature importance oriented interpretations. This approach, in addition to being used as a meta-explanation technique, can be used as an evaluation or selection tool for multiple feature importance techniques. Experimentation strongly indicates that an ensemble of multiple interpretation techniques yields considerably more truthful explanations.