Toward Explainable Multiparty Learning: A Contrastive Knowledge Sharing Framework.

Multiparty learning provides solutions for training joint models with decentralized data under legal and practical constraints. However, traditional multiparty learning approaches are confronted with obstacles, such as system heterogeneity, statistical heterogeneity, and incentive design. Determining how to deal with these challenges and further improve the efficiency and performance of multiparty learning has become an urgent problem to be solved. In this article, we propose a novel contrastive multiparty learning framework for knowledge refinement and sharing with an accountable incentive mechanism. Since the existing parameter averaging method is contradictory to the learning paradigm of neural networks, we simulate the process of human cognition and communication and analogize multiparty learning as a many-to-one knowledge-sharing problem. The approach is capable of integrating the acquired explicit knowledge of each client in a transparent manner without privacy disclosure, and it reduces the dependence on data distribution and communication environments. The proposed scheme achieves significant improvement in model performance in a variety of scenarios, as we demonstrated through experiments on several real-world datasets.

Device-Performance-Driven Heterogeneous Multiparty Learning for Arbitrary Images.

Multiparty learning (MPL) is an emerging framework for privacy-preserving collaborative learning. It enables individual devices to build a knowledge-shared model and remaining sensitive data locally. However, with the continuous increase of users, the heterogeneity gap between data and equipment becomes wider, which leads to the problem of model heterogeneous. In this article, we concentrate on two practical issues: data heterogeneous problem and model heterogeneous problem, and propose a novel personal MPL method named device-performance-driven heterogeneous MPL (HMPL). First, facing the data heterogeneous problem, we focus on the problem of various devices holding arbitrary data sizes. We introduce a heterogeneous feature-map integration method to adaptively unify the various feature maps. Meanwhile, to handle the model heterogeneous problem, as it is essential to customize models for adapting to the various computing performances, we propose a layer-wise model generation and aggregation strategy. The method can generate customized models based on the device's performance. In the aggregation process, the shared model parameters are updated through the rules that the network layers with the same semantics are aggregated with each other. Extensive experiments are conducted on four popular datasets, and the result demonstrates that our proposed framework outperforms the state of the art (SOTA).

Heterogeneous Multi-Party Learning With Data-Driven Network Sampling.

Multi-party learning provides an effective approach for training a machine learning model, e.g., deep neural networks (DNNs), over decentralized data by leveraging multiple decentralized computing devices, subjected to legal and practical constraints. Different parties, so-called local participants, usually provide heterogenous data in a decentralized mode, leading to non-IID data distributions across different local participants which pose a notorious challenge for multi-party learning. To address this challenge, we propose a novel heterogeneous differentiable sampling (HDS) framework. Inspired by the dropout strategy in DNNs, a data-driven network sampling strategy is devised in the HDS framework, with differentiable sampling rates which allow each local participant to extract from a common global model the optimal local model that best adapts to its own data properties so that the size of the local model can be significantly reduced to enable more efficient inference. Meanwhile, co-adaptation of the global model via learning such local models allows for achieving better learning performance under non-IID data distributions and speeds up the convergence of the global model. Experiments have demonstrated the superiority of the proposed method over several popular multi-party learning techniques in the multi-party settings with non-IID data distributions.

Flexible cellulose-based piezoelectric composite membrane involving PVDF and BaTiO3 synthesized with the assistance of TEMPO-oxidized cellulose nanofibrils.

High-performance flexible barium titanate (BaTiO3)-based piezoelectric devices have gained much attention. However, it is still a challenge to prepare flexible polymer/BaTiO3-based composite materials with uniform distribution and high performance due to the high viscosity of polymers. In this study, novel hybrid BaTiO3 particles were synthesized with assistance of TEMPO-oxidized cellulose nanofibrils (CNFs) via a low-temperature hydrothermal method and explored for their application in piezoelectric composites. Specifically, Ba2+ was adsorbed on uniformly dispersed CNFs with a large amount of negative charge on their surface, which nucleated, resulting in the synthesis of evenly dispersed CNF-BaTiO3. The obtained CNF-BaTiO3 possessed a uniform particle size, few impurities, high crystallinity and dispersity, high compatibility with the polymer substrate and surface activity due to the existence of CNFs. Subsequently, both polyvinylidene fluoride (PVDF) and TEMPO-oxidized CNFs were employed as piezoelectric substrates for the fabrication of a CNF/PVDF/CNF-BaTiO3 composite membrane with a compact structure, displaying the tensile strength of 18.61 ± 3.75 MPa and elongation at break of 3.06 ± 1.33%. Finally, a thin piezoelectric generator (PEG) was assembled, which output a considerable open-circuit voltage (4.4 V) and short-circuit current (200 nA), and could also power a light-emitting diode and charge a 1 µF capacitor to 3.66 V in 500 s. Its longitudinal piezoelectric constant (d 33) was 5.25 ± 1.04 pC N-1 even with a small thickness. It also exhibited high sensitivity to human movement, outputting a voltage of about 9 V and current of 739 nA for only a footstep. Thus, it exhibited good sensing property and energy harvesting property, presenting practical application prospects. This work provides a new idea for the preparation of hybrid BaTiO3 and cellulose-based piezoelectric composite materials.

Anisotropic Wooden Electromechanical Transduction Devices Enhanced by TEMPO Oxidization and PDMS.

In order to increase the number and contact probability of electric dipole on cellulose, acid and alkali treatment was employed to extract hemicellulose and lignin from original wood to gain a highly oriented cellulose frame. The combined means with 2,2,6,6-tetramethylpiperidine-1-oxyl-NaBr-NaClO oxidation and impregnation of PDMS with compression was subsequently used to enhance its mechanical performance and electromechanical conversion. The assembled wooden electromechanical device (10 mm × 10 mm × 1 mm) exhibits the maximum open-circuit voltage (V OC) of 11.75 V and short-circuit current (I SC) of 211.01 nA as stepped by foot. It can be sliced to fabricate a flexible sensor with high sensitivity displaying V OC of 2.88 V and I SC of 210.09 nA under the tapped state. Its highly oriented wood fiber makes it display significant anisotropy in terms of mechanical and electromechanical performance for multidirectional sense. This strategy will exactly provide reference for developing other high-performance piezoelectric devices.

MGAT: Multi-view Graph Attention Networks.

Multi-view graph embedding is aimed at learning low-dimensional representations of nodes that capture various relationships in a multi-view network, where each view represents a type of relationship among nodes. Multitudes of existing graph embedding approaches concentrate on single-view networks, that can only characterize one simple type of proximity relationships among objects. However, most of the real-world complex systems possess multiple types of relationships among entities. In this paper, a novel approach of graph embedding for multi-view networks is proposed, named Multi-view Graph Attention Networks (MGAT). We explore an attention-based architecture for learning node representations from each single view, the network parameters of which are constrained by a novel regularization term. In order to collaboratively integrate multiple types of relationships in different views, a view-focused attention method is explored to aggregate the view-wise node representations. We evaluate the proposed algorithm on several real-world datasets, and it demonstrates that the proposed approach outperforms existing state-of-the-art baselines.