Feature importance estimates that inform users about the degree to which given inputs influence the output of a predictive model are crucial for understanding, validating, and interpreting machine-learning models. However, providing fast and accurate estimates of feature importance for high-dimensional data, and quantifying the uncertainty of such estimates remain open challenges. Here, we frame the task of providing explanations for the decisions of machine-learning models as a causal learning task, and train causal explanation (CXPlain) models that learn to estimate to what degree certain inputs cause outputs in another machine-learning model. CXPlain can, once trained, be used to explain the target model in little time, and enables the quantification of the uncertainty associated with its feature importance estimates via bootstrap ensembling. We present experiments that demonstrate that CXPlain is significantly more accurate and faster than existing model-agnostic methods for estimating feature importance. In addition, we confirm that the uncertainty estimates provided by CXPlain ensembles are strongly correlated with their ability to accurately estimate feature importance on held-out data.

Large language models (LLMs) exhibit remarkable capabilities in visual inspection of medical time-series data, achieving proficiency comparable to human clinicians. However, their broad scope limits domain-specific precision, and proprietary weights hinder fine-tuning for specialized datasets. In contrast, small specialized models (SSMs) excel in targeted tasks but lack the contextual reasoning required for complex clinical decision-making. To address these challenges, we propose ConMIL (Conformalized Multiple Instance Learning), a decision-support SSM that integrates seamlessly with LLMs. By using Multiple Instance Learning (MIL) to identify clinically significant signal segments and conformal prediction for calibrated set-valued outputs, ConMIL enhances LLMs' interpretative capabilities for medical time-series analysis. Experimental results demonstrate that ConMIL significantly improves the performance of state-of-the-art LLMs, such as ChatGPT4.0 and Qwen2-VL-7B. Specifically, \ConMIL{}-supported Qwen2-VL-7B achieves 94.92% and 96.82% precision for confident samples in arrhythmia detection and sleep staging, compared to standalone LLM accuracy of 46.13% and 13.16%. These findings highlight the potential of ConMIL to bridge task-specific precision and broader contextual reasoning, enabling more reliable and interpretable AI-driven clinical decision support.

Most of the responses were quite clarifying (esp. in regards to noting that there is a separate Omega for each x!) and I'll increase my score from a (4) to a (5). I still think much more clarity is needed in describing the methodology, the overall goal, and defining carefully what they mean by "causal." Other referees noted that the interest in this paper is Granger causality and not in understanding what might have happened under intervention (the authors mention this as well in their response, noting that "our goal is not to estimate what would happen if a particular feature's value changed"). In the present form, I worry that most readers will think the paper is about this second notion and get quite confused. They do this by framing the question causally: i.e. identify which inputs causally affect the outputs of a machine learning model.

This is a solid, well-supported addition to the literature on explainable methods for neural networks. Evaluations of the proposed system show faster and higher quality results than related systems. Reviewers found the author response compelling.

While machine learning offers diverse techniques suitable for exploring various medical research questions, a cohesive synergistic framework can facilitate the integration and understanding of new approaches within unified model development and interpretation. We therefore introduce the Medical Artificial Intelligence Toolbox (MAIT), an explainable, open-source Python pipeline for developing and evaluating binary classification, regression, and survival models on tabular datasets. MAIT addresses key challenges (e.g., high dimensionality, class imbalance, mixed variable types, and missingness) while promoting transparency in reporting (TRIPOD+AI compliant). Offering automated configurations for beginners and customizable source code for experts, MAIT streamlines two primary use cases: Discovery (feature importance via unified scoring, e.g., SHapley Additive exPlanations - SHAP) and Prediction (model development and deployment with optimized solutions). Moreover, MAIT proposes new techniques including fine-tuning of probability threshold in binary classification, translation of cumulative hazard curves to binary classification, enhanced visualizations for model interpretation for mixed data types, and handling censoring through semi-supervised learning, to adapt to a wide set of data constraints and study designs. We provide detailed tutorials on GitHub, using four open-access data sets, to demonstrate how MAIT can be used to improve implementation and interpretation of ML models in medical research.

Feature importance estimates that inform users about the degree to which given inputs influence the output of a predictive model are crucial for understanding, validating, and interpreting machine-learning models. However, providing fast and accurate estimates of feature importance for high-dimensional data, and quantifying the uncertainty of such estimates remain open challenges. Here, we frame the task of providing explanations for the decisions of machine-learning models as a causal learning task, and train causal explanation (CXPlain) models that learn to estimate to what degree certain inputs cause outputs in another machine-learning model. CXPlain can, once trained, be used to explain the target model in little time, and enables the quantification of the uncertainty associated with its feature importance estimates via bootstrap ensembling. We present experiments that demonstrate that CXPlain is significantly more accurate and faster than existing model-agnostic methods for estimating feature importance. In addition, we confirm that the uncertainty estimates provided by CXPlain ensembles are strongly correlated with their ability to accurately estimate feature importance on held-out data.

Neural network have achieved remarkable successes in many scientific fields. However, the interpretability of the neural network model is still a major bottlenecks to deploy such technique into our daily life. The challenge can dive into the non-linear behavior of the neural network, which rises a critical question that how a model use input feature to make a decision. The classical approach to address this challenge is feature attribution, which assigns an important score to each input feature and reveal its importance of current prediction. However, current feature attribution approaches often indicate the importance of each input feature without detail of how they are actually processed by a model internally. These attribution approaches often raise a concern that whether they highlight correct features for a model prediction. For a neural network model, the non-linear behavior is often caused by non-linear activation units of a model. However, the computation behavior of a prediction from a neural network model is locally linear, because one prediction has only one activation pattern. Base on the observation, we propose an instance-wise linearization approach to reformulates the forward computation process of a neural network prediction. This approach reformulates different layers of convolution neural networks into linear matrix multiplication. Aggregating all layers' computation, a prediction complex convolution neural network operations can be described as a linear matrix multiplication $F(x) = W \cdot x + b$. This equation can not only provides a feature attribution map that highlights the important of the input features but also tells how each input feature contributes to a prediction exactly. Furthermore, we discuss the application of this technique in both supervise classification and unsupervised neural network learning parametric t-SNE dimension reduction.

Machine learning (ML) is a focal element of digitization that affects many areas of modern society: besides driving a plethora of physical and virtual products already woven into our daily lives, such as smartphones and social media platforms, ML techniques can be leveraged to power a wide range of business applications [1, 2]. Although ML as an umbrella term comprises various techniques, some of which are aimed at different purposes, most ML algorithms are designed to calculate empirical predictions based on given data [2]. This prediction-oriented approach to ML is widely referred to as supervised learning, predictive analytics, or predictive modeling, and initially requires at least two data sets: one for model training and one for testing [2, 3]. While the former allows a given ML algorithm to "learn" patterns that connect the model input and output, the latter serves to evaluate the predictive accuracy of a trained model. In practice, if a corresponding ML model is attributed to possess a sufficient degree of predictive power, it may be deployed in a productive environment to compute real-world predictions, e.g., to support managerial decision making. The application of supervised learning in business contexts is highly relevant as it may drive applications in the fields of predictive maintenance, financial fraud detection, personalized product recommendation, and more. Consequently, the global ML market size was valued at US$ 34.56 billion in 2021 and is expected to grow to US$ 74.99 billion by 2028 at a compound annual growth rate of 25.7% [4]. Given the enormous business potential of ML, a considerable number of companies have already begun to launch data analytics initiatives to automate their processes or support their decision making over the last years.

In the context of Just-In-Time Software Defect Prediction (JIT-SDP), Concept drift (CD) can occur due to changes in the software development process, the complexity of the software, or changes in user behavior that may affect the stability of the JIT-SDP model over time. Additionally, the challenge of class imbalance in JIT-SDP data poses a potential risk to the accuracy of CD detection methods if rebalancing is implemented. This issue has not been explored to the best of our knowledge. Furthermore, methods to check the stability of JIT-SDP models over time by considering labeled evaluation data have been proposed. However, it should be noted that future data labels may not always be available promptly. We aim to develop a reliable JIT-SDP model using CD point detection directly by identifying changes in the interpretation of unlabeled simplified and resampled data. To evaluate our approach, we first obtained baseline methods based on model performance monitoring to identify CD points on labeled data. We then compared the output of the proposed methods with baseline methods based on performance monitoring of threshold-dependent and threshold-independent criteria using well-known performance measures in CD detection methods, such as accuracy, MDR, MTD, MTFA, and MTR. We also utilize the Friedman statistical test to assess the effectiveness of our approach. As a result, our proposed methods show higher compatibility with baseline methods based on threshold-independent criteria when applied to rebalanced data, and with baseline methods based on threshold-dependent criteria when applied to simple data.

Saliency methods provide post-hoc model interpretation by attributing input features to the model outputs. Current methods mainly achieve this using a single input sample, thereby failing to answer input-independent inquiries about the model. We also show that input-specific saliency mapping is intrinsically susceptible to misleading feature attribution. Current attempts to use 'general' input features for model interpretation assume access to a dataset containing those features, which biases the interpretation. Addressing the gap, we introduce a new perspective of input-agnostic saliency mapping that computationally estimates the high-level features attributed by the model to its outputs. These features are geometrically correlated, and are computed by accumulating model's gradient information with respect to an unrestricted data distribution. To compute these features, we nudge independent data points over the model loss surface towards the local minima associated by a human-understandable concept, e.g., class label for classifiers. With a systematic projection, scaling and refinement process, this information is transformed into an interpretable visualization without compromising its model-fidelity. The visualization serves as a stand-alone qualitative interpretation. With an extensive evaluation, we not only demonstrate successful visualizations for a variety of concepts for large-scale models, but also showcase an interesting utility of this new form of saliency mapping by identifying backdoor signatures in compromised classifiers.