Explaining the Behavior of Black-Box Prediction Algorithms with Causal Learning
Sani Analytics；Department of Biostatistics, Columbia University（生物统计学系，哥伦比亚大学）；Department of Computer Science, Johns Hopkins University（计算机科学系，约翰霍普金斯大学）
Numair Sani, Daniel Malinsky, Ilya Shpitser
发表信息：
Transactions on Machine Learning Research (05/2025)
Causal approaches to post-hoc explainability for black-box prediction models (e.g., deep neural networks trained on image pixel data) have become increasingly popular. However, existing approaches have two important shortcomings: (i) the "explanatory units" are micro-level inputs into the relevant prediction model, e.g., image pixels, rather than interpretable macro-level features that are more useful for understanding how to possibly change the algorithm's behavior, and (ii) existing approaches assume there exists no unmeasured confounding between features and target model predictions, which fails to hold when the explanatory units are macro-level variables. Our focus is on the important setting where the analyst has no access to the inner workings of the target prediction algorithm, rather only the ability to query the output of the model in response to a particular input. To provide causal explanations in such a setting, we propose to learn causal graphical representations that allow for arbitrary unmeasured confounding among features. We demonstrate the resulting graph can differentiate between interpretable features that causally influence model predictions versus those that are merely associated with model predictions due to confounding. Our approach is motivated by a counterfactual theory of causal explanation wherein good explanations point to factors that are "difference-makers" in an interventionist sense.
Although deep learning models have been successfully applied to a variety of tasks, due to the millions of parameters, they are becoming increasingly opaque and complex. In order to establish trust for their widespread commercial use, it is important to formalize a principled framework to reason over these models. In this work, we use ideas from causal inference to describe a general framework to reason over CNN models. Specifically, we build a Structural Causal Model (SCM) as an abstraction over a specific aspect of the CNN. We also formulate a method to quantitatively rank the filters of a c…
Black box machine learning models are currently being used for high stakes decision-making throughout society, causing problems throughout healthcare, criminal justice, and in other domains. People have hoped that creating methods for explaining these black box models will alleviate some of these problems, but trying to \textit{explain} black box models, rather than creating models that are \textit{interpretable} in the first place, is likely to perpetuate bad practices and can potentially cause catastrophic harm to society. There is a way forward -- it is to design models that are inherently …
Understanding how a learned black box works is of crucial interest for the future of Machine Learning. In this paper, we pioneer the question of the global interpretability of learned black box models that assign numerical values to symbolic sequential data. To tackle that task, we propose a spectral algorithm for the extraction of weighted automata (WA) from such black boxes. This algorithm does not require the access to a dataset or to the inner representation of the black box: the inferred model can be obtained solely by querying the black box, feeding it with inputs and analyzing its outpu…
Model explanations based on pure observational data cannot compute the effects of features reliably, due to their inability to estimate how each factor alteration could affect the rest. We argue that explanations should be based on the causal model of the data and the derived intervened causal models, that represent the data distribution subject to interventions. With these models, we can compute counterfactuals, new samples that will inform us how the model reacts to feature changes on our input. We propose a novel explanation methodology based on Causal Counterfactuals and identify the limit…
Ludwig Maximilian University（路德维希-马克西米利安大学）；Courant Institute of Mathematical Sciences, New York University（纽约大学柯朗数学科学研究所）
We present the Rate-Distortion Explanation (RDE) framework, a mathematically well-founded method for explaining black-box model decisions. The framework is based on perturbations of the target input signal and applies to any differentiable pre-trained model such as neural networks. Our experiments demonstrate the framework's adaptability to diverse data modalities, particularly images, audio, and physical simulations of urban environments.
Carnegie Mellon University（卡内基梅隆大学）；Petuum Inc.（Petuum公司）；Arizona State University（亚利桑那州立大学）
Interpretable machine learning has gained much attention recently. Briefness and comprehensiveness are necessary in order to provide a large amount of information concisely when explaining a black-box decision system. However, existing interpretable machine learning methods fail to consider briefness and comprehensiveness simultaneously, leading to redundant explanations. We propose the variational information bottleneck for interpretation, VIBI, a system-agnostic interpretable method that provides a brief but comprehensive explanation. VIBI adopts an information theoretic principle, informati…
We introduce a variable importance measure to quantify the impact of individual input variables to a black box function. Our measure is based on the Shapley value from cooperative game theory. Many measures of variable importance operate by changing some predictor values with others held fixed, potentially creating unlikely or even logically impossible combinations. Our cohort Shapley measure uses only observed data points. Instead of changing the value of a predictor we include or exclude subjects similar to the target subject on that predictor to form a similarity cohort. Then we apply Shapl…
Algorithms are powerful and necessary tools behind a large part of the information we use every day. However, they may introduce new sources of bias, discrimination and other unfair practices that affect people who are unaware of it. Greater algorithm transparency is indispensable to provide more credible and reliable services. Moreover, requiring developers to design transparent algorithm-driven applications allows them to keep the model accessible and human understandable, increasing the trust of end users. In this paper we present EBAnO, a new engine able to produce prediction-local explana…
How can we understand classification decisions made by deep neural networks? Many existing explainability methods rely solely on correlations and fail to account for confounding, which may result in potentially misleading explanations. To overcome this problem, we define the Causal Concept Effect (CaCE) as the causal effect of (the presence or absence of) a human-interpretable concept on a deep neural net's predictions. We show that the CaCE measure can avoid errors stemming from confounding. Estimating CaCE is difficult in situations where we cannot easily simulate the do-operator. To mitigat…
By their nature, the composition of black box models is opaque. This makes the ability to generate explanations for the response to stimuli challenging. The importance of explaining black box models has become increasingly important given the prevalence of AI and ML systems and the need to build legal and regulatory frameworks around them. Such explanations can also increase trust in these uncertain systems. In our paper we present RICE, a method for generating explanations of the behaviour of black box models by (1) probing a model to extract model output examples using sensitivity analysis; …
With the increasing adoption of predictive models trained using machine learning across a wide range of high-stakes applications, e.g., health care, security, criminal justice, finance, and education, there is a growing need for effective techniques for explaining such models and their predictions. We aim to address this problem in settings where the predictive model is a black box; That is, we can only observe the response of the model to various inputs, but have no knowledge about the internal structure of the predictive model, its parameters, the objective function, and the algorithm used t…
School of Electrical & Computer Engineering（电气与计算机工程学院）；Georgia Institute of Technology（佐治亚理工学院）
We develop a method for generating causal post-hoc explanations of black-box classifiers based on a learned low-dimensional representation of the data. The explanation is causal in the sense that changing learned latent factors produces a change in the classifier output statistics. To construct these explanations, we design a learning framework that leverages a generative model and information-theoretic measures of causal influence. Our objective function encourages both the generative model to faithfully represent the data distribution and the latent factors to have a large causal influence o…
We interpret the predictions of any black-box structured input-structured output model around a specific input-output pair. Our method returns an "explanation" consisting of groups of input-output tokens that are causally related. These dependencies are inferred by querying the black-box model with perturbed inputs, generating a graph over tokens from the responses, and solving a partitioning problem to select the most relevant components. We focus the general approach on sequence-to-sequence problems, adopting a variational autoencoder to yield meaningful input perturbations. We test our meth…
Language Technology Institute（语言技术研究所）；Carnegie Mellon University（卡内基梅隆大学）
Explaining underlying causes or effects about events is a challenging but valuable task. We define a novel problem of generating explanations of a time series event by (1) searching cause and effect relationships of the time series with textual data and (2) constructing a connecting chain between them to generate an explanation. To detect causal features from text, we propose a novel method based on the Granger causality of time series between features extracted from text such as N-grams, topics, sentiments, and their composition. The generation of the sequence of causal entities requires a co…
University of Pisa（比萨大学）；ISTI-CNR（意大利国家研究委员会信息科学与技术研究所）
In the last years many accurate decision support systems have been constructed as black boxes, that is as systems that hide their internal logic to the user. This lack of explanation constitutes both a practical and an ethical issue. The literature reports many approaches aimed at overcoming this crucial weakness sometimes at the cost of scarifying accuracy for interpretability. The applications in which black box decision systems can be used are various, and each approach is typically developed to provide a solution for a specific problem and, as a consequence, delineating explicitly or impli…
Tata Consultancy Services Limited（塔塔咨询服务有限公司）；Indian Institute of Technology Kharagpur（印度理工学院卡哈拉格普尔分校）；Aristotle University of Thessaloniki（塞萨洛尼基亚里士多德大学）
Multi-party dialogues are common in enterprise social media on technical as well as non-technical topics. The outcome of a conversation may be positive or negative. It is important to analyze why a dialogue ends with a particular sentiment from the point of view of conflict analysis as well as future collaboration design. We propose an explainable time series mining algorithm for such analysis. A dialogue is represented as an attributed time series of occurrences of keywords, EMPATH categories, and inferred sentiments at various points in its progress. A special decision tree, with decision me…
In attempts to produce ML models less reliant on spurious patterns in NLP datasets, researchers have recently proposed curating counterfactually augmented data (CAD) via a human-in-the-loop process in which given some documents and their (initial) labels, humans must revise the text to make a counterfactual label applicable. Importantly, edits that are not necessary to flip the applicable label are prohibited. Models trained on the augmented data appear, empirically, to rely less on semantically irrelevant words and to generalize better out of domain. While this work draws loosely on causal th…
Institute for Artificial Intelligence, Tsinghua University（清华大学人工智能研究院）；University of Florida（佛罗里达大学）；Weill Cornell Medicine（威尔康奈尔医学院）
Algorithmic fairness has aroused considerable interests in data mining and machine learning communities recently. So far the existing research has been mostly focusing on the development of quantitative metrics to measure algorithm disparities across different protected groups, and approaches for adjusting the algorithm output to reduce such disparities. In this paper, we propose to study the problem of identification of the source of model disparities. Unlike existing interpretation methods which typically learn feature importance, we consider the causal relationships among feature variables …
Deep neural networks are complex and opaque. As they enter application in a variety of important and safety critical domains, users seek methods to explain their output predictions. We develop an approach to explaining deep neural networks by constructing causal models on salient concepts contained in a CNN. We develop methods to extract salient concepts throughout a target network by using autoencoders trained to extract human-understandable representations of network activations. We then build a bayesian causal model using these extracted concepts as variables in order to explain image class…
Explaining Black-box Model Predictions via Two-level Nested Feature Attributions with Consistency Property
STAIR Lab, Chiba Institute of Technology（千叶技术学院STAIR实验室）；School of Mathematics and Statistics, The University of Melbourne（墨尔本大学数学与统计学学院）；ZOZO Research（ZOZO研究）
Temporal Graph Neural Network (TGNN) has been receiving a lot of attention recently due to its capability in modeling time-evolving graph-related tasks. Similar to Graph Neural Networks, it is also non-trivial to interpret predictions made by a TGNN due to its black-box nature. A major approach tackling this problems in GNNs is by analyzing the model' responses on some perturbations of the model's inputs, called perturbation-based explanation methods. While these methods are convenient and flexible since they do not need internal access to the model, does this lack of internal access prevent t…
University of Cagliari（卡利亚里大学）；Pluribus One（Pluribus One公司）
Machine-learning models have been recently used for detecting malicious Android applications, reporting impressive performances on benchmark datasets, even when trained only on features statically extracted from the application, such as system calls and permissions. However, recent findings have highlighted the fragility of such in-vitro evaluations with benchmark datasets, showing that very few changes to the content of Android malware may suffice to evade detection. How can we thus trust that a malware detector performing well on benchmark data will continue to do so when deployed in an oper…
Lenovo Research（联想研究院）；University of Chinese Academy of Sciences（中国科学院大学）
Despite outstanding contribution to the significant progress of Artificial Intelligence (AI), deep learning models remain mostly black boxes, which are extremely weak in explainability of the reasoning process and prediction results. Explainability is not only a gateway between AI and society but also a powerful tool to detect flaws in the model and biases in the data. Local Interpretable Model-agnostic Explanation (LIME) is a recent approach that uses an interpretable model to form a local explanation for the individual prediction result. The current implementation of LIME adopts the linear r…
University of Massachusetts Amherst（马萨诸塞大学阿默斯特分校）；University of California, San Diego（加州大学圣迭戈分校）
There has been a recent resurgence of interest in explainable artificial intelligence (XAI) that aims to reduce the opaqueness of AI-based decision-making systems, allowing humans to scrutinize and trust them. Prior work in this context has focused on the attribution of responsibility for an algorithm's decisions to its inputs wherein responsibility is typically approached as a purely associational concept. In this paper, we propose a principled causality-based approach for explaining black-box decision-making systems that addresses limitations of existing methods in XAI. At the core of our fr…
Peter the Great St.Petersburg Polytechnic University（彼得大帝圣彼得堡理工大学）
A new robust algorithm based of the explanation method SurvLIME called SurvLIME-KS is proposed for explaining machine learning survival models. The algorithm is developed to ensure robustness to cases of a small amount of training data or outliers of survival data. The first idea behind SurvLIME-KS is to apply the Cox proportional hazards model to approximate the black-box survival model at the local area around a test example due to the linear relationship of covariates in the model. The second idea is to incorporate the well-known Kolmogorov-Smirnov bounds for constructing sets of predicted …
Fudan University（复旦大学）；Tencent Weixin Group（腾讯微信事业群）；Microsoft Research Asia（微软亚洲研究院）；Arizona State University（亚利桑那州立大学）
Risk scoring systems have been widely deployed in many applications, which assign risk scores to users according to their behavior sequences. Though many deep learning methods with sophisticated designs have achieved promising results, the black-box nature hinders their applications due to fairness, explainability, and compliance consideration. Rule-based systems are considered reliable in these sensitive scenarios. However, building a rule system is labor-intensive. Experts need to find informative statistics from user behavior sequences, design rules based on statistics and assign weights to…
We introduce a new model-agnostic explanation technique which explains the prediction of any classifier called CLE. CLE gives an faithful and interpretable explanation to the prediction, by approximating the model locally using an interpretable model. We demonstrate the flexibility of CLE by explaining different models for text, tabular and image classification, and the fidelity of it by doing simulated user experiments.
University of Southern California（南加州大学）；Facebook AI（脸书人工智能研究院）
Recommendation is a prevalent application of machine learning that affects many users; therefore, it is important for recommender models to be accurate and interpretable. In this work, we propose a method to both interpret and augment the predictions of black-box recommender systems. In particular, we propose to interpret feature interactions from a source recommender model and explicitly encode these interactions in a target recommender model, where both source and target models are black-boxes. By not assuming the structure of the recommender system, our approach can be used in general setti…
Rutgers University（罗格斯大学）；Renmin University of China（中国人民大学）
State-of-the-art recommender systems have the ability to generate high-quality recommendations, but usually cannot provide intuitive explanations to humans due to the usage of black-box prediction models. The lack of transparency has highlighted the critical importance of improving the explainability of recommender systems. In this paper, we propose to extract causal rules from the user interaction history as post-hoc explanations for the black-box sequential recommendation mechanisms, whilst maintain the predictive accuracy of the recommendation model. Our approach firstly achieves counterfac…
University of Tsukuba（筑波大学）；Riken AIP（日本理化学研究所革新智能综合研究中心）
We aim to explain a black-box classifier with the form: `data X is classified as class Y because X \textit{has} A, B and \textit{does not have} C' in which A, B, and C are high-level concepts. The challenge is that we have to discover in an unsupervised manner a set of concepts, i.e., A, B and C, that is useful for the explaining the classifier. We first introduce a structural generative model that is suitable to express and discover such concepts. We then propose a learning process that simultaneously learns the data distribution and encourages certain concepts to have a large causal influenc…
Modern deep learning models for NLP are notoriously opaque. This has motivated the development of methods for interpreting such models, e.g., via gradient-based saliency maps or the visualization of attention weights. Such approaches aim to provide explanations for a particular model prediction by highlighting important words in the corresponding input text. While this might be useful for tasks where decisions are explicitly influenced by individual tokens in the input, we suspect that such highlighting is not suitable for tasks where model decisions should be driven by more complex reasoning.…
Although interactive learning puts the user into the loop, the learner remains mostly a black box for the user. Understanding the reasons behind queries and predictions is important when assessing how the learner works and, in turn, trust. Consequently, we propose the novel framework of explanatory interactive learning: in each step, the learner explains its interactive query to the user, and she queries of any active classifier for visualizing explanations of the corresponding predictions. We demonstrate that this can boost the predictive and explanatory powers of and the trust into the learn…
TimeSAE: Causal Sparse Decoding for Faithful Explanations of Black-Box Time Series Models
Telecom Paris（电信巴黎大学）；Institut Polytechnique de Paris（巴黎理工学院）；Technical University of Munich（慕尼黑技术大学）；Helmholtz Munich（海德堡-慕尼黑马克斯·普朗克研究所）；Munich Center for Machine Learning（慕尼黑机器学习中心）；Munich Data Science Institute（慕尼黑数据科学研究所）
While deep reinforcement learning has proven to be successful in solving control tasks, the "black-box" nature of an agent has received increasing concerns. We propose a prototype-based post-hoc policy explainer, ProtoX, that explains a blackbox agent by prototyping the agent's behaviors into scenarios, each represented by a prototypical state. When learning prototypes, ProtoX considers both visual similarity and scenario similarity. The latter is unique to the reinforcement learning context, since it explains why the same action is taken in visually different states. To teach ProtoX about vis…
Stanford School of Engineering（斯坦福大学工程学院）；Stanford School of Medicine（斯坦福大学医学院）
The causal explanation of image misclassifications is an understudied niche, which can potentially provide valuable insights in model interpretability and increase prediction accuracy. This study trains CIFAR-10 on six modern CNN architectures, including VGG16, ResNet50, GoogLeNet, DenseNet161, MobileNet V2, and Inception V3, and explores the misclassification patterns using conditional confusion matrices and misclassification networks. Two causes are identified and qualitatively distinguished: morphological similarity and non-essential information interference. The former cause is not model d…
Peter the Great St.Petersburg Polytechnic University（彼得大帝圣彼得堡理工大学）
A new modification of the explanation method SurvLIME called SurvLIME-Inf for explaining machine learning survival models is proposed. The basic idea behind SurvLIME as well as SurvLIME-Inf is to apply the Cox proportional hazards model to approximate the black-box survival model at the local area around a test example. The Cox model is used due to the linear relationship of covariates. In contrast to SurvLIME, the proposed modification uses $L_{\infty }$-norm for defining distances between approximating and approximated cumulative hazard functions. This leads to a simple linear programming pr…
Does Location Matter? A Black-Box Causal Audit Reveals Model-Dependent Audio Grounding in LALM Hallucination
Sharif University of Technology（谢里夫理工大学）；Amirkabir University of Technology（阿米尔卡比尔理工大学）；New Uzbekistan University（新乌兹别克斯坦大学）；Okinawa Institute of Science and Technology（冲绳科学技术大学院大学）
Simon Fraser University（西蒙菲莎大学）；Columbia University（哥伦比亚大学）
Obtaining an explanation for an SQL query result can enrich the analysis experience, reveal data errors, and provide deeper insight into the data. Inference query explanation seeks to explain unexpected aggregate query results on inference data; such queries are challenging to explain because an explanation may need to be derived from the source, training, or inference data in an ML pipeline. In this paper, we model an objective function as a black-box function and propose BOExplain, a novel framework for explaining inference queries using Bayesian optimization (BO). An explanation is a predic…
