High-throughput omics data present challenges for binary classification due to platform variability, batch effects, missing values, and high dimensionality. This study presents a novel Rank-Based Learning (RBL) method that leverages relative feature rankings to improve robustness and generalizability. We evaluated RBL against established methods like Logistic Regression (LR) and Random Forest (RF) using simulated data and two real-world plasma proteomics datasets: early-stage small cell lung cancer (SCLC) and duodenopancreatic neuroendocrine tumors (dpNET) in patients with Multiple Endocrine Neoplasia type 1 (MEN1). In simulation experiments, RBL outperformed LR under conditions involving batch effects, missing data, and varying numbers of true differential features. In SCLC, RBL yielded a test AUC of 0.76 (95% CI: 0.42-1.00), surpassing LR with Lasso (0.65 [95% CI: 0.47-0.84]) and RF with feature importance (0.59 [95% CI: 0.33-0.87]). In dpNET, RBL achieved an AUC of 0.83 (95% CI: 0.67-0.97) on the development set and 0.80 (95% CI: 0.54-0.98) on the test set, outperforming LR with Lasso (0.57 [95% CI: 0.40-0.77]) and RF with feature importance (0.53 [95% CI: 0.29-0.77]). By emphasizing feature ranking rather than absolute expression levels, RBL effectively mitigates the impact of non-biological variation. Overall, RBL improves the predictive accuracy of diagnostic models for complex diseases and provides a promising framework for developing more reliable and generalizable diagnostic tools from omics data, moving them closer to clinical application.

Mission-critical applications are of significant importance to sixth generation (6G)’s massive and ubiquitous Internet of things (IoT) communications. The mission-critical applications mostly fall within the scope of finite blocklength (FBL), and in order to assess the information freshness, age of information (AoI) has been introduced. However, packet error is inevitable in the FBL regime, which exerts impacts on the time for successful packet transmission, and thus increases the AoI. To optimize the performance of AoI, the management of queue packets is an effective way. Motivated by optimizing the AoI performance in the FBL regime, we consider a system equipped with a single buffer, and propose two schemes of packet management in this article. We subsequently derive the closed-form expressions for the average AoI and the average peak AoI and we discuss the relationship between AoI and the factors, i.e. the blocklength, data generation rate and signal-to-noise ratio. Afterwards, we give the optimal blocklength expression associated with the optimal AoI. In order to examine the information timeliness in the network under the proposed schemes, the closed-form expressions of the average delay are deduced. The simulation results validate the theoretical analysis and demonstrate the advantage of the proposed scheme in terms of the performance of AoI, delay, and their trade-off.

We consider a robust energy-efficient reconfigurable intelligent surface (RIS)-aided multi-antenna decode-and-forward (DF) relay cooperative multiple-input multiple-output (MIMO). Although RIS and relay share some similarities in common, they have fundamental differences and can indeed complement each other. Due to the passive characteristic of RIS, it is much challenging to obtain the perfect channel state information (CSI) and the channel estimation error (CEE) is inevitable in practice. Taking into account the imperfect CSI, we formulate the robust energy efficiency (EE) optimization problems under the bounded CEE and statistical CEE models, where the precoding matrices at the source and relay, and the passive beamforming at the RIS in two slots are jointly designed. At first, the original problems under two CEE models are transformed into deterministic forms with the help of S-procedure and Bernstein-type Inequality, respectively. Subsequently, the reformulated problems are solved by the alternating optimization (AO)-based Dinkelbach algorithm in an iterative manner. Particularly, for the passive beamforming subproblem, the semi-definite relaxation (SDR) method and penalty concave-convex procedure (PCCP) method are utilized to deal with the rank-one constraint. Numerical simulations demonstrate that the EE performance of the considered scheme obviously outperforms the benchmarks. Simulation results also show the superiorities of the robust EE optimization compared with the non-robust optimization.

We consider an energy-efficient reconfigurable intelligent surface (RIS)-aided multi-antenna decode-and-forward (DF) relay cooperative multiple input multiple output (MIMO). Although RIS and relay share some similarities in common, they have fundamental differences and can indeed complement each other. Due to the passive characteristic of RIS, it is much challenging to obtain the perfect channel state information (CSI) and the channel estimation error (CEE) is inevitable. Aiming to the robust energy efficiency (EE) optimization of the considered system with the norm-bounded CEE, we formulate the robust EE optimization problem where the precoding matrices at source and relay as well as the phase shifts of RIS are jointly optimized with the worst-case minimum-rate and total transmit power constraints. The original problem is firstly transformed into a deterministic problem with the help of S-procedure, then an alternating optimization (AO)-based Dinkelbach algorithm is applied to solve it. Numerical simulation results demonstrate the superiorities of the robust EE optimization compared with the non-robust scheme, and also show the validity and good convergency of the proposed algorithm. Furthermore, the results also illustrate the effects of various factors on the performance.

Abstract The purpose of vehicle reidentification is to distinguish the same vehicle from different vehicles using different cameras. The main challenge of this task is the significant intra‐instance discrepancy of the same vehicle from different views and the subtle inter‐instance differences of similar vehicles from the same views. To address this problem, researchers have attempted to align features from different views, such as using additional metadata (colour, type, key points, mask, etc.) to improve performance. Although these attempts improve the performance of vehicle reidentification, considerable efforts are required to create additional precise annotations to the data before using these methods. This results in expensive data preparation costs, rendering these methods less convenient to use. Therefore, we propose a novel deep learning network, called a mixed‐attention‐based regional soft partition network, to address this problem. This network does not require additional metadata annotations; it only trains the identity label as a supervision signal and uses soft partition attention to identify specific vehicle regions. Experiments showed that the performance of the proposed method was comparable to that of the state‐of‐the‐art method with additional annotations on the VeRri‐776 and VERI‐Wild datasets.