The importance of 30-day patients’ readmissions (PRs) to intensive unit care stems from the significant cost and mortality risk when the patient’s chosen class (i.e., readmitted or not to the hospital) is incorrect. The overall accuracy (OA) of the PRs classification obtained in the literature is still moderate, particularly for machine learning (ML)-enabled ANNs, where OA is around 65%, resulting in 35% critical wrong decisions. To improve such an OA, a three-stage ML-assisted algorithm employing both support vector machines (SVMs) and artificial neural networks (ANNs) techniques is proposed. Starting with a well-fitted PR accuracies distribution and using mathematical modeling, the global optimal accuracies’ interval of misclassified patients (OIMP) is obtained, and a novel theorem for the generalized secretary problem is introduced and used to select the most likely well-classified patients. The algorithm’s final phase consists of flipping the remaining incorrectly classified patients in the OIMP. Using the presented approach with ANN and SVM, the OA was raised by 5% and 19%, respectively. The proposed approach may be used for any binary classification application and expanded to any multi-class problem.

Deep Neural Networks (DDNs) have shown vulnerability to well-designed adversarial examples. Researchers in industry and academia have proposed many adversarial example defense techniques. However, none can provide complete robustness. The cutting-edge defense techniques offer partial reliability. Thus, complementing them with another layer of protection is a must, especially for mission-critical applications. This paper proposes a novel Online Selection and Relabeling Algorithm (OSRA) that opportunistically utilizes a limited number of crowdsourced workers to maximize the ML system’s robustness. OSRA strives to use crowdsourced workers effectively by selecting the most suspicious inputs and moving them to the crowdsourced workers to be validated and corrected. As a result, the impact of adversarial examples gets reduced, and accordingly, the ML system becomes more robust. We also proposed a heuristic threshold selection method that contributes to enhancing the prediction system’s reliability. We empirically validated our proposed algorithm and found that it can efficiently and optimally utilize the allocated budget for crowdsourcing. It is also effectively integrated with a state-of-the-art black-box defense technique, resulting in a more robust system. Simulation results show that OSRA can outperform a random selection algorithm by 60% and achieve comparable performance to an optimal offline selection benchmark. They also show that OSRA’s performance has a positive correlation with system robustness.

p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 9.0px Helvetica} span.s1 {font: 7.5px Helvetica} xAbstract—In this paper, the physical layer security of a dualhop underlay uplink cognitive radio network is investigated over Nakagami-m fading channels. Specifically, multiple secondary sources are taking turns in accessing the licensed spectrum of the primary users and communicating with a multiantenna secondary base station (D) through the aid of a multiantenna relay R in the presence of M eavesdroppers that are also equipped with multiple antennas. Among the remaining nodes, one jammer is randomly selected to transmit an artificial noise to disrupt all the eavesdroppers that are attempting to intercept the communication of the legitimate links i.e., Si -R and R-D. The received signals at each node are combined using maximum-ratio combining. Secrecy analysis is provided by deriving closed-form and asymptotic expressions for the secrecy outage probability. The impact of several key parameters on the system’s secrecy e.g., transmit power of the sources, number of eavesdroppers, maximum tolerated interference power, and the number of diversity branches is investigated. Importantly, by considering two scenarios, namely (i) absence and (ii) presence of a friendly jammer, new insights are obtained for the considered communication system. Especially, we tend to answer to the following question: Can better secrecy be achieved without jamming by considering a single antenna at eavesdroppers and multiple-ones at the legitimate users (i.e., relay and enduser) rather than sending permanently an artificial noise and considering that both the relay and the destination are equipped with a single antenna, while multiple antennas are used by the eavesdroppers? The obtained results are corroborated through Monte Carlo simulation and show that the system’s security can be enhanced by adjusting the aforementioned parameters.