Search for: All records

Despite some promising results in federated learning using game-theoretical methods, most existing studies mainly employ a one-level game in either a cooperative or competitive environment, failing to capture the complex dynamics among participants in practice. To address this issue, we propose DualGFL, a novel federated learning framework with a dual-level game in cooperative-competitive environments. DualGFL includes a lower-level hedonic game where clients form coalitions and an upper-level multi-attribute auction game where coalitions bid for training participation.At the lower-level DualGFL, we introduce a new auction-aware utility function and propose a Pareto-optimal partitioning algorithm to find a Pareto-optimal partition based on clients' preference profiles.At the upper-level DualGFL, we formulate a multi-attribute auction game with resource constraints and derive equilibrium bids to maximize coalitions' winning probabilities and profits. A greedy algorithm is proposed to maximize the utility of the central server.Extensive experiments on real-world datasets demonstrate DualGFL's effectiveness in improving both server utility and client utility. 

Cell membranes are incredibly complex environments containing hundreds of components. Despite substantial advances in the past decade, fundamental questions related to lipid-lipid interactions and heterogeneity persist. This review explores the complexity of lipid membranes, showcasing recent advances in vibrational spectroscopy to characterize the structure, dynamics, and interactions at the membrane interface. We include an overview of modern techniques such as surface-enhanced infrared spectroscopy as a steady-state technique with single-bilayer sensitivity, two-dimensional sum-frequency generation spectroscopy, and two-dimensional infrared spectroscopy to measure time-evolving structures and dynamics with femtosecond time resolution. Furthermore, we discuss the potential of multiscale molecular dynamics (MD) simulations, focusing on recently developed simulation algorithms, which have emerged as a powerful approach to interpret complex spectra. We highlight the ongoing challenges in studying heterogeneous environments in multicomponent membranes via current vibrational spectroscopic techniques and MD simulations. Overall, this review provides an up-to-date comprehensive overview of the powerful combination of vibrational spectroscopy and simulations, which has great potential to illuminate lipid-lipid, lipid-protein, and lipid-water interactions in the intricate conformational landscape of cell membranes. 

Abstract Actin, an important component of eukaryotic cell cytoskeleton, regulates cell shape and transport. The morphology and biochemical properties of actin filaments are determined by their structure and protein-protein contacts. Crowded environments can organize filaments into bundles, but less is known about how they affect F-actin structure. This study used 2D IR spectroscopy and spectral calculations to examine how crowding and bundling impact the secondary structure and local environments in filaments and weakly or strongly bundled networks. The results reveal that bundling induces changes in actin’s secondary structure, leading to a decrease inβ-sheet and an increase in loop conformations. Strongly bundled networks exhibit a decrease in backbone solvent exposure, with less perturbedα-helices and nearly “locked”β-sheets. Similarly, the loops become less hydrated but maintain a dynamic environment. These findings highlight the role of loop structure in actin network morphology and stability under morphology control by PEG. 

The conduction mechanism of a family of high concentration lithium electrolytes (HCEs) is investigated. It is found in all HCEs that the molecular motions are regulated by the anion size and correlated to the HCE ionic resistivity. From the results, a mechanism involving highly correlated ionic networks is derived.