MOTIVATION: Accurate survival prediction is crucial for personalized cancer treatment but remains challenging for rare cancers due to limited data. Most deep learning models require large training datasets, which are unavailable for rare cancer types, creating a significant clini-cal bottleneck. RESULTS: We propose MoESurv, a zero-sample survival prediction framework that leverages a mix-ture-of-experts architecture to extract generalizable prognostic patterns from pan-cancer data. The model integrates shared experts, cancer-specific experts, and routing experts within an autoencoder to disentangle common and type-specific survival features. Evaluated on seven rare TCGA cancer types, MoESurv achieved state-of-the-art performance, improving the average C-index by 4 percentage points over the best baseline. Further-more, external validation across diverse populations and independent cohorts-including a Chinese glioma cohort (CGGA mRNAseq_693, C-index = 0.7433), a rare GBM IDH-mutant subtype (C-index = 0.8064), and the pan-cancer PCAWG cohort (C-index = 0.7090)-demonstrated that MoESurv possesses the most robust predictive per-formance, highlighting its generalizability. MoESurv also effectively stratified high- and low-risk patient groups and identified potential survival-associated genes, demonstrating both clinical utility and biological interpretability. AVAILABILITY: The code is freely available at https://github.com/HuaYC666/MoESurv and https://zenodo.org/records/20785500.

Adaptive knowledge transfer from pretrained source models to unlabeled target domains has emerged as a key paradigm in the Industrial Internet of Things. Such frameworks effectively mitigate distribution discrepancies between domains and ensure data privacy in cross-condition fault diagnosis. However, most existing methods require extensive parameter tuning for each source backbone, resulting in high computational costs, especially as the number of source domains or the complexity of models increases. Moreover, current methods typically learn the transferable weights of each source model at domain-level, inevitably causing instance-level bias. Thus, we propose a new idea of Specificity and Consistency Ensemble (SpeConE) for efficient multiple-source fault diagnosis, which avoids parameter tuning for each source backbone. SpeConE balances instance-specific and domain-consistent weights by jointly considering feature-feature and feature-output similarities. By fine-tuning the source bottlenecks guided by the specificity and consistency weights, SpeConE effectively achieves multi-source-free cross-domain adaptation. Additionally, a novel pseudo-label smoothing strategy is introduced to prevent overfitting in the target domain. Experiments show that, when adopting Swin Transformer, SpeConE outperforms similar methods (e.g., ∼+3% on single-device diagnosis and ∼+10% on cross-device diagnosis) while tuning only about 5% of the parameters.

Abstract Current adaptive torque balancing control of electric wheel-driven vehicle has shortcomings in electronic differential control of drive motor by using rotation speed mode. In order to solve this problem, an adaptive electronic differential control method of electric wheel-driven vehicle is proposed in this paper by torque balance. Firstly, by starting from the kinematics and dynamics of vehicle steering, the speed and force of each drived wheel in the steering are analyzed to explain the auxiliary role of electronic differential control to adaptive torque balancing, as well as influence of steering radius in vehicle. Then, electronic differential distribution by torque of wheel is used to control the abnormal jump interference and calculate the optimal combination of parameters in electronic differential control system. Finally, based on the optimal combination of these parameters, an adaptive electronic differential control of electric wheel-driven vehicle by torque balance is realized with fuzzy control in active and the reactive power. Experimental results show that the proposed method suppresses the abnormal jump interference factors of electronic differential control, as well as realizes the differential functions in control system. It has far-reaching significance by provideing a basic guarantee to realize adaptive electronic differential control system in electronic wheel-driven vehicle.

Summary This paper focuses on designing and implementing parallel Moving Least Squares (MLS) interpolation algorithms by exploiting the Graphics Processing Unit (GPU) for the usage in Meshfree methods. The MLS method is an approach for scattered points' approximation / interpolation, which is commonly employed as the shape functions in various Meshfree methods. To improve the computational efficiency in building stiffness matrices in Meshfree methods, we are specifically interested in parallelizing the MLS interpolation on the GPU. The accelerated Meshfree methods can be employed to numerically analyze large deformations of soil and rock masses such as the landslides. In this paper, we first introduce our previously proposed method for finding the k nearest neighboring points located within the local region of the interest point in MLS. We then develop one sequential and three parallel implementations for each of three variations of the MLS interpolation, including the serial implementation, the parallel implementation on the multi‐core CPU, the parallel implementation on a single GPU, and the parallel implementation on multi‐GPUs. To evaluate the computational performance of our GPU implementations, five groups of benchmark tests are conducted in both two‐dimensions and three‐dimensions. We observe that our GPU implementations can achieve satisfied speedups over the sequential CPU implementation for varied sizes of testing data. To benefit the community, all source code and testing data related to the presented parallel MLS interpolation are publicly available.

K* is rotamerically ensemble-based approach to compute the binding constant. However, its time-consuming feature limited its application. We present a novel algorithm that not only computes the partition function efficiently, but also avoids the exponential growth of execution time by iteratively pruning the sequence space until the sequence with highest affinity is identified.

This paper brings together methods from two different disciplines: statistics and machine learning. We address the problem of estimating the variance of cross-validation (CV) estimators of the generalization error. In particular, we approach the problem of variance estimation of the CV estimators of generalization error as a problem in approximating the moments of a statistic. The approximation illustrates the role of training and test sets in the performance of the algorithm. It provides a unifying approach to evaluation of various methods used in obtaining training and test sets and it takes into account the variability due to different training and test sets. For the simple problem of predicting the sample mean and in the case of smooth loss functions, we show that the variance of the CV estimator of the generalization error is a function of the moments of the random variables Y=Card(Sj ∩ Sj') and Y*=Card(Sjc ∩ Sj'c), where Sj, Sj' are two training sets, and Sjc, Sj'c are the corresponding test sets. We prove that the distribution of Y and Y* is hypergeometric and we compare our estimator with the one proposed by Nadeau and Bengio (2003). We extend these results in the regression case and the case of absolute error loss, and indicate how the methods can be extended to the classification case. We illustrate the results through simulation.

This paper investigates the problem of specifying and implementing physical boundary conditions for the Monte Carlo (MC) simulation of electron dynamics in semiconductor devices. The goal of this work is to establish an accurate and efficient ohmic boundary condition scheme for use in characterizing realistic device structures. In this work, three distinct physical models for specifying the boundary electrons at the ideal ohmic contacts of an N/sup +/-N-N/sup +/ GaAs Ballistic diode structure are investigated. This study demonstrates that a displaced Maxwellian scheme, which allows for an electron ensemble with momentum space displacement and random spread, presents definite computational advantages when one is interested in resolving asymmetries in the electron distribution function throughout the semiconductor device structure.>