Recently, deep autoencoder-like non-negative matrix factorization methodologies have achieved impressive performance in multi-view clustering. These methods encode the input data into latent features and decode the latent features to recover the input data, thus obtaining meaningful features. They primarily focus on exploring the complementary information of multi-view data, but overlook that the features reconstructing the input data well may not be discriminative because the quality of the input data is typically not good enough. For example, the input data usually contains noise. To address this issue, we propose an A symmetric D eep A utoencoder-like N on-negative M atrix F actorization for Multi-view Clustering (ADA-NMF). The framework recovers the underlying data from the input data using a deep autoencoder-like non-negative matrix factorization model while leveraging l 1 -norm regularization to explicitly model and mitigate noise. To this end, we devise an asymmetric deep autoencoder-like architecture that decouples the encoder and decoder components, thereby enabling independent optimization of the encoding and decoding processes. This asymmetric design enhances the capacity of the model to discriminatively extract semantic features, facilitating precise reconstruction of the underlying data. Following the semantic features, ADA-NMF further adaptively learns view-specific local similarity graphs and derives a low-rank tensor representation for multi-view data to capture the cross-view consistency and complementary information. Finally, an efficient optimization algorithm is designed to tackle the optimization problem. Our primary contribution to the field of artificial intelligence is the proposal of a novel deep non-negative matrix factorization framework for multi-view clustering, with demonstrated applicability to engineering tasks involving noisy multi-modal data.

This paper focuses on scalability and robustness of spectral clustering for extremely large-scale datasets with limited resources. Two novel algorithms are proposed, namely, ultra-scalable spectral clustering (U-SPEC) and ultra-scalable ensemble clustering (U-SENC). In U-SPEC, a hybrid representative selection strategy and a fast approximation method for K-nearest representatives are proposed for the construction of a sparse affinity sub-matrix. By interpreting the sparse sub-matrix as a bipartite graph, the transfer cut is then utilized to efficiently partition the graph and obtain the clustering result. In U-SENC, multiple U-SPEC clusterers are further integrated into an ensemble clustering framework to enhance the robustness of U-SPEC while maintaining high efficiency. Based on the ensemble generation via multiple U-SEPC's, a new bipartite graph is constructed between objects and base clusters and then efficiently partitioned to achieve the consensus clustering result. It is noteworthy that both U-SPEC and U-SENC have nearly linear time and space complexity, and are capable of robustly and efficiently partitioning 10-million-level nonlinearly-separable datasets on a PC with 64 GB memory. Experiments on various large-scale datasets have demonstrated the scalability and robustness of our algorithms. The MATLAB code and experimental data are available at https://www.researchgate.net/publication/330760669.

Our concern is nonlinear clustering on large-scale dataset. While existing popular kernels (RBF, Polynomials, Spatial Pyramid, etc.) are popularly used for implicitly mapping data into a high-dimensional or infinite dimensional space in order to generalise linear clustering methods, using these kernels cannot make kernel clustering approaches directly applicable for large scale dataset, since large scale kernel matrix or similarity matrix consumes a lot of memory (e.g., 7,450 GB memory over 1 million samples of data). To solve this problem, we introduce an Euler clustering approach. Euler clustering employs Euler kernels in order to intrinsically map the input data onto a complex space of the same dimension as the input or twice, so that Euler clustering can get rid of kernel trick and does not need to rely on any approximation or random sampling on kernel function/matrix, whilst performing a more robust nonlinear clustering against noise and outliers. Moreover, since the original Euler kernel cannot generate a non-negative similarity matrix and thus is inapplicable to spectral clustering, we introduce a positive Euler kernel, and more importantly we have proved when it can generate a non-negative similarity matrix. We apply Euler kernel and the proposed positive Euler kernel to kernel k-means and spectral clustering so as to develop Euler k-means and Euler spectral clustering, respectively. An efficient Stiefel-manifold-based gradient method and an equivalent weighted positive Euler k-means are derived for fast computation of Euler spectral clustering and further alleviating the impact of discretization of the cluster membership indicators in Euler spectral clustering. The results show that the proposed Euler clustering approach achieves overall better clustering performance compared to using popular Mercer kernels and approximation models, whilst keeping the computational complexity of the same magnitude as the most popular linear clustering method k-means.

Automated annotation of protein function is challenging. As the number of sequenced genomes rapidly grows, the vast majority of proteins can only be annotated computationally. Nature often brings several domains together to form multi-domain and multi-functional proteins with a vast number of possibilities, and each domain may fulfill its own function independently or in a concerted manner with its neighbors. Thus, it is evident that the protein function prediction problem is naturally and inherently Multi-Instance Multi-Label (MIML) learning tasks. Based on the state-of-the-art MIML algorithm MIMLNN, we propose a novel ensemble MIML learning framework EnMIMLNN and design three algorithms for this task by combining the advantage of three kinds of Hausdorff distance metrics. Experiments on seven real-world organisms covering the biological three-domain system, i.e., archaea, bacteria, and eukaryote, show that the EnMIMLNN algorithms are superior to most state-of-the-art MIML and Multi-Label learning algorithms.

The recognition of microRNA (miRNA)-binding residues in proteins is helpful to understand how miRNAs silence their target genes. It is difficult to use existing computational method to predict miRNA-binding residues in proteins due to the lack of training examples. To address this issue, unlabeled data may be exploited to help construct a computational model. Semisupervised learning deals with methods for exploiting unlabeled data in addition to labeled data automatically to improve learning performance, where no human intervention is assumed. In addition, miRNA-binding proteins almost always contain a much smaller number of binding than nonbinding residues, and cost-sensitive learning has been deemed as a good solution to the class imbalance problem. In this work, a novel model is proposed for recognizing miRNA-binding residues in proteins from sequences using a cost-sensitive extension of Laplacian support vector machines (CS-LapSVM) with a hybrid feature. The hybrid feature consists of evolutionary information of the amino acid sequence (position-specific scoring matrices), the conservation information about three biochemical properties (HKM) and mutual interaction propensities in protein-miRNA complex structures. The CS-LapSVM receives good performance with an F1 score of 26.23 ± 2.55% and an AUC value of 0.805 ± 0.020 superior to existing approaches for the recognition of RNA-binding residues. A web server called SARS is built and freely available for academic usage.