RESUMO
Self-supervised clustering has garnered widespread attention due to its ability to discover latent clustering structures without the need for external labels. However, most existing approaches on self-supervised clustering lack of inherent interpretability in the data clustering process. In this paper, we propose a differentiable self-supervised clustering method with intrinsic interpretability (DSC2I), which provides an interpretable data clustering mechanism by reformulating clustering process based on differentiable programming. To be specific, we first design a differentiable mutual information measurement to explicitly train a neural network with analytical gradients, which avoids variational inference and learns a discriminative and compact representation. Then, an interpretable clustering mechanism based on differentiable programming is devised to transform fundamental clustering process (i.e., minimum intra-cluster distance, maximum inter-cluster distance) into neural networks and convert cluster centers to learnable neural parameters, which allows us to obtain a transparent and interpretable clustering layer. Finally, a unified optimization method is designed, in which the differentiable representation learning and interpretable clustering can be optimized simultaneously in a self-supervised manner. Extensive experiments demonstrate the effectiveness of the proposed DSC2I method compared with 16 clustering approaches.
RESUMO
Starch is a biomass polymer material with a high yield and comprehensive source. It is used as a raw material for preparing adhesives because of its highly active hydroxyl group. However, poor adhesion and water resistance hinder the application of starch-based adhesives (SBAs). Based on this, the starch was modified through graft copolymerization with itaconic acid as a cross-linking agent, methyl methacrylate and methyl acrylate as copolymers. Additionally, reed fibers were synergistically modified with polydopamine deposition to prepare an environmentally friendly SBA used in plywood production. Fourier transform infrared spectroscopy (FT-IR), nuclear magnetic resonance spectroscopy (1H NMR), X-ray diffraction (XRD), and thermogravimetric analysis (TG) demonstrate that copolymerization of methyl methacrylate and methyl acrylate with starch improves the shear strength, water resistance, and thermal stability of the SBA. Compared to unmodified starch, the modified SBA exhibits a 129 % increase in dry strength and achieves a wet strength of 1.36 MPa. Fukui function, Frontier orbit theory, and molecular dynamics simulation have shown that itaconic acid promotes the copolymerization of starch and acrylate monomers. The modified starch has fewer hydrogen bonds, less order, and a denser macromolecular network structure, which provides a reference for studying the molecular interaction mechanisms of SBAs.
Assuntos
Acrilatos , Simulação de Dinâmica Molecular , Amido , Succinatos , Amido/química , Adesivos/química , Espectroscopia de Infravermelho com Transformada de Fourier , Água/química , MetacrilatosRESUMO
A 65-year-old man underwent excision of a giant mesenteric fibromatosis (MF) via combined splenectomy and partial transverse colectomy. Pathological examination confirmed the presence of MF, whereas genetic testing indicated that the tumor was sensitive to tamoxifen. Over a 1-year follow-up, no symptoms of abdominal discomfort or recurrence was noted.