arXiv subjects

Junwen Xu

Publications and source records attributed to Junwen Xu.

4 recordsLinked to original sources

arXiv2024

An Evolutionary Network Architecture Search Framework with Adaptive Multimodal Fusion for Hand Gesture Recognition

Hand gesture recognition (HGR) based on multimodal data has attracted considerable attention owing to its great potential in applications. Various manually designed multimodal deep networks have performed well in multimodal HGR (MHGR), but most of existing algorithms require a lot of expert experience and time-consuming manual trials. To address these issues, we propose an evolutionary network architecture search framework with the adaptive multimodel fusion (AMF-ENAS). Specifically, we design an encoding space that simultaneously considers fusion positions and ratios of the multimodal data, allowing for the automatic construction of multimodal networks with different architectures through decoding. Additionally, we consider three input streams corresponding to intra-modal surface electromyography (sEMG), intra-modal accelerometer (ACC), and inter-modal sEMG-ACC. To automatically adapt to various datasets, the ENAS framework is designed to automatically search a MHGR network with appropriate fusion positions and ratios. To the best of our knowledge, this is the first time that ENAS has been utilized in MHGR to tackle issues related to the fusion position and ratio of multimodal data. Experimental results demonstrate that AMF-ENAS achieves state-of-the-art performance on the Ninapro DB2, DB3, and DB7 datasets.

cs.CV↗

arXiv2024

Multiple Population Alternate Evolution Neural Architecture Search

The effectiveness of Evolutionary Neural Architecture Search (ENAS) is influenced by the design of the search space. Nevertheless, common methods including the global search space, scalable search space and hierarchical search space have certain limitations. Specifically, the global search space requires a significant amount of computational resources and time, the scalable search space sacrifices the diversity of network structures and the hierarchical search space increases the search cost in exchange for network diversity. To address above limitation, we propose a novel paradigm of searching neural network architectures and design the Multiple Population Alternate Evolution Neural Architecture Search (MPAE), which can achieve module diversity with a smaller search cost. MPAE converts the search space into L interconnected units and sequentially searches the units, then the above search of the entire network be cycled several times to reduce the impact of previous units on subsequent units. To accelerate the population evolution process, we also propose the the population migration mechanism establishes an excellent migration archive and transfers the excellent knowledge and experience in the migration archive to new populations. The proposed method requires only 0.3 GPU days to search a neural network on the CIFAR dataset and achieves the state-of-the-art results.

cs.NE↗

arXiv2022

Exciton-Coupled Coherent Magnons in a 2D Semiconductor

Two-dimensional (2D) magnetic semiconductors feature both tightly-bound excitons with large oscillator strength and potentially long-lived coherent magnons due to the presence of bandgap and spatial confinement. While magnons and excitons are energetically mismatched by orders of magnitude, their coupling can lead to efficient optical access to spin information. Here we report strong magnon-exciton coupling in the 2D van der Waals (vdW) antiferromagnetic (AFM) semiconductor CrSBr. Coherent magnons launched by above-gap excitation modulate the interlayer hybridization, which leads to dynamic modulation of excitonic energies. Time-resolved exciton sensing reveals magnons that can coherently travel beyond 7 micrometer, with coherence time above 5 ns. We observe this exciton-coupled coherent magnons in both even and odd number of layers, with and without compensated magnetization, down to the bilayer limit. Given the versatility of vdW heterostructures, these coherent 2D magnons may be basis for optically accessible magnonics and quantum interconnects.

cond-mat.mes-hall↗

arXiv2019

Electrical control of Majorana Bound States Using Magnetic Stripes

A hybrid semiconductor-superconductor nanowire on the top of a magnetic film in the stripe phase experiences a magnetic texture from the underlying fringing fields. The Zeeman interaction with the highly inhomogeneous magnetic textures generates a large synthetic spin-orbit coupling. We show that this platform can support the formation of Majorana bound states (MBS) localized at the ends of the nanowire. The transition to the topological superconducting phase not only depends on the nanowire parameters and stripe size but also on the relative orientation of the stripes with respect to the nanowire axis. Topological phase transitions with the corresponding emergence or destruction of MBS can be induced by reorienting the stripes or shifting their position, which can be achieved by passing a charge current through the magnetic film or by applying electrically-controlled strain to it. The proposed platform removes the need for external magnetic fields and offers a non-invasive electrical tuning of MBS with the perturbation (current or strain) acting only on thee magnetic film.

cond-mat.supr-con↗