We introduce a new survival tree method for censored failure time data that incorporates three key advancements over traditional approaches. First, we develop a more computationally efficient splitting procedure that effectively mitigates the end-cut preference problem, and we propose an intersected validation strategy to reduce the variable selection bias inherent in greedy searches. Second, we present a novel framework for determining tree structures through fused regularization. In combination with conventional pruning, this approach enables the merging of non-adjacent terminal nodes, producing more parsimonious and interpretable models. Third, we address inference by constructing valid confidence intervals for median survival times within the subgroups identified by the final tree. To achieve this, we apply bootstrap-based bias correction to standard errors. The proposed method is assessed through extensive simulation studies and illustrated with data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) study.

Convolutional Neural Networks (CNNs) achieve high performance in image classification tasks but are challenging to deploy on resource-limited hardware due to their large model sizes. To address this issue, we leverage Mutual Information, a metric that provides valuable insights into how deep learning models retain and process information through measuring the shared information between input features or output labels and network layers. In this study, we propose a structured filter-pruning approach for CNNs that identifies and selectively retains the most informative features in each layer. Our approach successively evaluates each layer by ranking the importance of its feature maps based on Conditional Mutual Information (CMI) values, computed using a matrix-based Rényi α-order entropy numerical method. We propose several formulations of CMI to capture correlation among features across different layers. We then develop various strategies to determine the cutoff point for CMI values to prune unimportant features. This approach allows parallel pruning in both forward and backward directions and significantly reduces model size while preserving accuracy. Tested on the VGG16 architecture with the CIFAR-10 dataset, the proposed method reduces the number of filters by more than a third, with only a 0.32% drop in test accuracy. Convolution Neural Network (CNN) has achieved remarkable success in various tasks such as image classification, object detection, and segmentation (Zhang et al., 2019), (Li et al., 2021). Deeper architectures such as VGG16 (Simonyan & Zisserman, 2014) and ResNet (He et al., 2016) have shown superior performance in handling complex image classification tasks. However, the effectiveness of these networks is often reliant on very deep and wide architectures, resulting in a very large number of parameters that lead to longer training and inference time, and create challenges when deploying them on resource-constrained devices (Blalock et al., 2020), (Yang et al., 2017).

Counterfactual questions are among the most important topics in business. I hear companies asking this kind of questions all the time. Afterward, the average user spending was 100 $. But how do we know what they would have spent if we didn't do our action?" These problems are usually addressed through A/B testing.

Multistage sequential decision-making scenarios are commonly seen in the healthcare diagnosis process. In this paper, an active learning-based method is developed to actively collect only the necessary patient data in a sequential manner. There are two novelties in the proposed method. First, unlike the existing ordinal logistic regression model which only models a single stage, we estimate the parameters for all stages together. Second, it is assumed that the coefficients for common features in different stages are kept consistent. The effectiveness of the proposed method is validated in both a simulation study and a real case study. Compared with the baseline method where the data is modeled individually and independently, the proposed method improves the estimation efficiency by 62\%-1838\%. For both simulation and testing cohorts, the proposed method is more effective, stable, interpretable, and computationally efficient on parameter estimation. The proposed method can be easily extended to a variety of scenarios where decision-making can be done sequentially with only necessary information.

This is the fourth in a series of articles intended to make Machine Learning more approachable to those without technical training. Prior articles introduce the concept of Machine Learning, show how the process of learning works in general, and describe commonly used algorithms. You can start the series here. In this installment of the series, we review some common measures and considerations when assessing the effectiveness and value of a classification algorithm. To get started, let's assume that an algorithm has been developed on a training set and has generalized well.

Assessing heterogeneous treatment effects has become a growing interest in advancing precision medicine. Individualized treatment effects (ITE) play a critical role in such an endeavor. Concerning experimental data collected from randomized trials, we put forward a method, termed random forests of interaction trees (RFIT), for estimating ITE on the basis of interaction trees (Su et al., 2009). To this end, we first propose a smooth sigmoid surrogate (SSS) method, as an alternative to greedy search, to speed up tree construction. RFIT outperforms the traditional `separate regression' approach in estimating ITE. Furthermore, standard errors for the estimated ITE via RFIT can be obtained with the infinitesimal jackknife method. We assess and illustrate the use of RFIT via both simulation and the analysis of data from an acupuncture headache trial.

Empirical performance models play an important role in the development of planning portfolios that make a per-domain or per-problem configuration of its search components. Even though such portfolios have shown their power when compared to other systems in current benchmarks, there is no clear evidence that they are capable to differentiate problems (instances) having similar input properties (in terms of objects, goals, etc.) but fairly different runtime for a given planner. In this paper we present a study of empirical performance models that are trained using problems having the same configuration, with the objective of guiding the models to recognize the underlying differences existing among homogeneous problems. In addition we propose a set of new features that boost the prediction capabilities under such scenarios. The results show that the learned models clearly performed over random classifiers, which reinforces the hypothesis that the selection of planners can be done on a per-instance basis when configuring a portfolio.