Federated Learning (FL) enables model training on distributed devices while preserving data privacy. However, malicious clients can submit fabricated model updates to fraudulently obtain training rewards, a behavior known as free-rider attacks. Existing detection-based solutions analyze anomalies in model updates but lack direct evidence of local training, making it fail to fully prevent free-riders. To address this limitation, we propose zkVFL, a verifiable FL framework leveraging Zero-Knowledge Proofs (ZKP) to ensure the integrity of local training while preserving privacy. To reduce the computational overhead of proof generation in ZKP, zkVFL introduces two novel techniques: (i) anomaly-aware client sampling to selectively perform ZKP verification and (ii) A recursive ZKP protocol (ReMPoT), incorporating a pruning-based layer selection technique, reduces proof generation costs. Experimental results demonstrate that zkVFL improves the accuracy and convergence of FL training under free-rider attacks while significantly reducing the computational and memory overhead of proof generation on resource-constrained devices.

A real-world system often exhibits complex dynamics arising from interaction among its subunits. In machine learning and data mining, these interactions are usually formulated as dependency and correlation among system variables. Similar to Convolution Neural Network dealing with spatially correlated features and Recurrent Neural Network with temporally correlated features, in this paper we present a novel deep learning model to tackle functionally interactive features by stacking a Conditional Restricted Boltzmann Machine and a Deep Neural Network (CRBM-DNN). Variables with their dependency relationships are organized into a bipartite graph, which is further converted into a Restricted Boltzmann Machine conditioned by domain knowledge. We integrate this CRBM and a DNN into one deep learning model constrained by one overall cost function. CRBM-DNN can solve both supervised and unsupervised learning problems. Compared to a regular neural network of the same size, CRBM-DNN has fewer parameters so they require fewer training samples. We perform extensive comparative studies with a large number of supervised learning and unsupervised learning methods using several challenging real-world datasets, and achieve significant superior performance.

This paper presents a new algorithm, Reinforced and Informed Network-based Clustering(RINC), for finding unknown groups of similar data objects in sparse and largely non-overlapping feature space where a network structure among features can be observed. Sparse and non-overlapping unlabeled data become increasingly common and available especially in text mining and biomedical data mining. RINC inserts a domain informed model into a modelless neural network. In particular, our approach integrates physically meaningful feature dependencies into the neural network architecture and soft computational constraint. Our learning algorithm efficiently clusters sparse data through integrated smoothing and sparse auto-encoder learning. The informed design requires fewer samples for training and at least part of the model becomes explainable. The architecture of the reinforced network layers smooths sparse data over the network dependency in the feature space. Most importantly, through back-propagation, the weights of the reinforced smoothing layers are simultaneously constrained by the remaining sparse auto-encoder layers that set the target values to be equal to the raw inputs. Empirical results demonstrate that RINC achieves improved accuracy and renders physically meaningful clustering results.

Recently, security problems in massive multiple-input multiple-output (MIMO) network draw more and more attention. In this paper, a security problem for a multi-pair massive MIMO two-way amplify-and-forward (AF) relay network is considered, where multiple users exchange confidential messages via a common relay equipped with large number of antennas. We propose an optimal power allocation scheme that maximizes the secrecy sum rate (SSR) under power constraints, when the multi-pair sources act as both potential eavesdroppers who intend to wiretap the information of other pairs and conventional users who exchange their information with pairings. Simulation results show that the proposed scheme outperforms the equal power allocation especially under the circumstance of large number of relay antennas. Furthermore, the number of user-pairs, the Signal to Noise Ratio (SNR) and the number of relay antennas are all factors which influence the performance of SSR in this network.

In the time-division-duplex (TDD) system, the channel state information (CSI) is mainly obtained through channel estimation. The active eavesdropper is able to attack the channel estimation phase by sending the same pilot signals as the legitimate users send to the base station without being detected, which can destroy the CSI acquired through channel estimation and greatly decrease the communication quality, so it is very necessary to detect the active eavesdropper before the data transmission. In this paper, we propose to design a mechanism to detect active eavesdropper in the massive MIMO network. Specifically, the sender will intentionally add noise to the pilot sequence so that it becomes harder for the bad guys to learn the pilot sequence. At the same time, if the real sender and the attacker signals arrive at the receiver simultaneously, the received signals will demonstrate special properties in the rank of the matrix. we design a mechanism to estimate the ranks to detect active eavesdropper. Comparison between the approach and MDL-based detector is conducted.

BACKGROUND: Stratification of patient subpopulations that respond favorably to treatment or experience and adverse reaction is an essential step toward development of new personalized therapies and diagnostics. It is currently feasible to generate omic-scale biological measurements for all patients in a study, providing an opportunity for machine learning models to identify molecular markers for disease diagnosis and progression. However, the high variability of genetic background in human populations hampers the reproducibility of omic-scale markers. In this paper, we develop a biological network-based regularized artificial neural network model for prediction of phenotype from transcriptomic measurements in clinical trials. To improve model sparsity and the overall reproducibility of the model, we incorporate regularization for simultaneous shrinkage of gene sets based on active upstream regulatory mechanisms into the model. RESULTS: We benchmark our method against various regression, support vector machines and artificial neural network models and demonstrate the ability of our method in predicting the clinical outcomes using clinical trial data on acute rejection in kidney transplantation and response to Infliximab in ulcerative colitis. We show that integration of prior biological knowledge into the classification as developed in this paper, significantly improves the robustness and generalizability of predictions to independent datasets. We provide a Java code of our algorithm along with a parsed version of the STRING DB database. CONCLUSION: In summary, we present a method for prediction of clinical phenotypes using baseline genome-wide expression data that makes use of prior biological knowledge on gene-regulatory interactions in order to increase robustness and reproducibility of omic-scale markers. The integrated group-wise regularization methods increases the interpretability of biological signatures and gives stable performance estimates across independent test sets.

A non-reused pilot (NRP) design scheme is presented for Large-scale multi-cell multiuser multiple-input multiple-output (LS-MIMO) systems. As the users at the cell edge suffer from severe pilot contamination (PC), the pilot design based on Chu sequences for two kinds of users is considered. The users in each cell are divided into center users and edge users according to the large-scale fading coefficients. The proposed pilot design scheme can reduce inter-cell interference of center users and improve the performance of edge users, whereby assigning phase shifted pilot groups to center users in adjacent cells and orthogonal pilot group to edge users in all cells.

This paper investigates the cooperative beamforming in the amplify-and-forward-based cognitive relay network. Beamforming coefficients are designed to maximize the sum rate of the secondary users with a total transmit power constraint of the relays and an interference power constraint at the primary receiver. The sum rate maximization problem is firstly solved by characterizing the achievable rate region. This method is optimal but of great complexity. Then, an iteration algorithm based on semidefinite programming (SDP) and the first order Taylor polynomial which has lower complexity but comparable performance is introduced. Simulation results confirm the feasibility of these two methods for the sum rate maximization problem in the cognitive relay network and demonstrate that the lower complexity method finds the global optimum of the problem just as the rate region method does.

In this paper, we propose a filter-and-forward (FF)beamforming strategy in the secondary relays for the one-way cognitive relay network with frequency selective fading channels. Our aim is to optimize the beamforming coefficients and the secondary transmitted power to maximize the SINR at the secondary destination, while satisfying the interference constraint on the primary network and the total relays' transmitted power restriction. To solve this problem, we propose a two-step iterative algorithm. Simulation results show that comparing the amplifyand-forward (AF) strategy, the filter-and-forward strategy (FF) can get higher SINR at the secondary destination by using the algorithm that we proposed.