Federated learning (FL) is vulnerable to backdoor attacks, where adversaries alter model behavior on target classification labels by embedding triggers into data samples. While these attacks have received considerable attention in horizontal FL, they are less understood for vertical FL (VFL), devices hold different features of the samples, and only server holds labels. In this work, we propose a novel attack VFL which (i) does not rely gradient information from (ii) considers potential...

Distributed matrix computations over large clusters can suffer from the problem of slow or failed worker nodes (called stragglers) which dominate overall job execution time. Coded computation utilizes concepts erasure coding to mitigate effect stragglers by running "coded" copies tasks comprising a job; are typically treated as erasures. While this is useful, there issues with applying, e.g., MDS codes in straightforward manner. Several practical scenarios involve sparse matrices. require...

In distributed computing systems, it is well recognized that worker nodes are slow (called stragglers) tend to dominate the overall job execution time. Coded computation utilizes concepts from erasure coding mitigate effect of stragglers by running “coded” copies tasks comprising a job. Stragglers typically treated as erasures in this process. While useful, there issues with applying, e.g., MDS codes straightforward manner. Specifically, several applications such matrix-vector products deal...

The current big data era routinely requires the processing of large-scale on massive distributed computing clusters. In these applications, sets are often so large that they cannot be housed in memory and/or disk any one computer. Thus, and typically across multiple nodes. Distributed computation is thus a necessity rather than luxury. widespread use such clusters presents several opportunities advantages over traditional paradigms. However, it also newer challenges where coding-theoretic...

Distributed computing systems are well-known to suffer from the problem of slow or failed nodes; these referred as stragglers. Straggler mitigation (for distributed matrix computations) has recently been investigated standpoint erasure coding in several works. In this work we present a strategy for matrix-vector multiplication based on convolutional coding. Our scheme can be decoded using lowcomplexity peeling decoder. The recovery process enjoys excellent numerical stability compared...

Distributed matrix computations - matrix-matrix or matrix-vector multiplications are well-recognized to suffer from the problem of stragglers (slow failed worker nodes). Much prior work in this area is (i) either sub-optimal terms its straggler resilience, (ii) suffers numerical problems, i.e., there a blow-up round-off errors decoded result owing high condition numbers corresponding decoding matrices. Our presents convolutional coding approach that removes these limitations. It optimal and...

The Phytophtora root and stem rot is a serious disease in soybean. It caused by the oomycete pathogen Phytophthora sojae. Growing resistant cultivars major method of controlling this disease. Resistance race- or gene-specific; single gene confers immunity against only subset P. sojae isolates. Unfortunately, rapid evolution new virulent pathotypes limits effectiveness an Rps ("resistance to sojae") 8-15 years. current study was designed investigate Rps12 set isolates using recombinant inbred...

In this paper, a statistical method for automatic detection of seizure and epilepsy in the dual-tree complex wavelet transform(DT-CWT) domain is proposed. Variances calculated from EEG signals their DT-CWT sub-bands are utilized as features classifiers such artificial neural network(ANN) support vector machine(SVM). Studies conducted using publicly available benchmark database to assess ability proposed number clinically relevant classification scenario which include healthy vs seizure,...

The overall execution time of distributed matrix computations is often dominated by slow worker nodes (stragglers) within the computation clusters. Recently, coding-theoretic techniques have been utilized to mitigate effect stragglers where are assigned job processing encoded submatrices original matrices. In many machine learning or optimization problems relevant matrices sparse. Several prior coded methods operate with dense linear combinations submatrices; this can significantly increase...

Coded matrix computation utilizes concepts from erasure coding to mitigate the effect of slow worker nodes (stragglers) in distributed setting. While this is useful, there are issues with applying, e.g., MDS codes a straightforward manner for problem. Several practical scenarios involve sparse matrices. typically require dense linear combinations submatrices original matrices which destroy their inherent sparsity; leads significantly higher times. Moreover, treating as erasures ignores...

In this paper, a comprehensive statistical analysis of electroencephalogram (EEG) signals is carried out in the dual tree complex wavelet transform domain using publicly available EEG database. It shown that variance and kurtosis can be effective distinguishing at sub-band levels. further parameters normal inverse Gaussian probability density function equally discriminate Thus, these quantities may used to characterize help researchers developing improved classifiers for detection epilepsy...

Fusarium virguliforme is a soil borne root pathogen that causes sudden death syndrome (SDS) in soybean [Glycine max (L.) Merrill]. Once the fungus invades xylem tissues, secretes toxins cause chlorosis and necrosis foliar tissues leading to defoliation, flower pod drop eventually of plants. Resistance F. partial governed by over 80 quantitative trait loci (QTL). We have conducted genome-wide association study (GWAS) for group 254 plant introductions lines using panel approximately 30,000...

In this paper, a comprehensive analysis for the discrimination of focal and non-focal electroencephalography (EEG) signals is carried out in ensemble empirical mode decomposition (EEMD) complete with adaptive noise (CEEMDAN) domains. A number spectral entropy-based features such as Shannon entropy, log-energy Renyi entropy Tsallis are calculated EEMD CEEMDAN domain EEG signals. lieu using from channels, differences between two adjacent channels used due to its robustness interference. The...

Radio access networks (RANs) in monolithic architectures have limited adaptability to supporting different network scenarios. Recently, open-RAN (O-RAN) techniques begun adding enormous flexibility RAN implementations. O-RAN is a natural architectural fit for cell-free massive multiple-input multiple-output (CFmMIMO) systems, where many geographically-distributed points (APs) are employed achieve ubiquitous coverage and enhanced user performance. In this paper, we address the decentralized...

In this work, a novel quaternary algebra has been employed to design encoders and decoders of any size. Both priority ordinary designed. The encoder in work may have arbitrary permutation input priorities; the procedure is generic works for such permutation. Although no ever done regarding arbitrarily large randomly prioritized domain, there exist numerous reports on smaller as well decoders. complexity present compared with some earlier our designs are more favorable terms complexity, especially

Supernumerary chromosome segments are known to harbor different transposons from their essential counterparts. The aim of this study was investigate the role in origin and evolution supernumerary asexual fungal pathogen Fusarium virguliforme. We compared genomes 11 isolates comprising six species that cause soybean sudden death syndrome (SDS) or bean root rot (BRR), identified significant levels genetic variation A+T-rich repeat blocks chromosomes A+T-neutral regions segments. were highly...

Federated learning (FL) is a popular technique for training global model on data distributed across client devices. Like other techniques, FL susceptible to straggler (slower or failed) clients. Recent work has proposed address this through device-to-device (D2D) offloading, which introduces privacy concerns. In paper, we propose novel straggler-optimal approach coded matrix computations can significantly reduce the communication delay and issues introduced from D2D transmissions in FL....

Straggler nodes are well-known bottlenecks of distributed matrix computations which induce reductions in computation/communication speeds. A common strategy for mitigating such stragglers is to incorporate MDS (maximum distance separable) codes into the framework; this can achieve resilience against an optimal number stragglers. However, these assign dense linear combinations submatrices workers increase non-zero entries encoded matrices, and adversely affect worker computation time. In...

Augmenting federated learning (FL) with direct device-to-device (D2D) communications can help improve conver-gence speed and reduce model bias through rapid local information exchange. However, data privacy concerns, device trust issues, unreliable wireless channels each pose challenges to determining an effective yet resource efficient D2D structure. In this paper, we develop a decentralized reinforcement (RL) methodology for graph discovery that promotes communication of non-sensitive...

In this paper, a statistical method has been proposed to identify motor imagery left and right hand movements from electroencephalogram (EEG) signals in the Dual Tree Complex Wavelet Transform (DTCWT) domain. The total experiment is carried out with publicly available benchmark BCI-competition 2003 Graz dataset. First, EEG are decomposed into several bands of real imaginary coefficients, then, some features like norm entropy standard deviation have calculated. From one way ANOVA analysis,...