基于深度胶囊网络融合模型的多声音事件检测
传统的胶囊网络架构是基于动态路由机制实现的，需要大量迭代和向量计算来更新权值系数，并且胶囊之间不存在信息共享，导致信息冗余。针对这一缺陷，本工作提出了一种基于融合深度胶囊网络的多声音事件检测模型，在门控卷积和3D卷积下通过动态路由减少了特征重叠导致的信息冗余，并且对原始特征进行编码，将其用于特征信息补充，提高了训练次数模型的速度和准确性。本工作使用 …
qingzhou-jiang
杨树国
正教授
教授，博士生导师，哈尔滨工业大学博士后。数据科学与信息技术研究中心主任，人工智能海洋技术场景化应用山东省工程研究中心主任，青岛市人工智能海洋技术创新中心主任，青岛科技大学数学与交叉研究院院长。美国佐治亚理工学院高级访问学者、香港中文大学高级访问学者、北京交通大学高级访问学者；山东省数学会常务理事、山东省应用统计学会常务理事、人工智能海洋学专业委员会常务委员。近年来，主持或参与国家自然科学基金、国防科工委、电子工业部、省自然基金、省重点科研计划、省高校科研计划、省优秀中青年科学家基金、青岛市科技发展计划项目等各级各类科研项目40多项。
•
•
1 分钟阅读时长
Design of Event-Triggered Adaptive Finite-Time Controller for Full-State Constrained Stochastic Nonlinear Systems with Unknown Control Directions
In this article, the problem of event-triggered (ET) adaptive finite-time tracking control is investigated for full-state constrained stochastic nonlinear systems with unknown …
zhongkun-feng
Attention-Enhanced Deep Learning Model for Reconstruction and Downscaling of Thermocline Depth in the Tropical Indian Ocean
Accurate estimation of high-resolution thermocline depth is important for investigating ocean processes and climate variability on multiple scales. Due to the sparse coverage and …
zhongkun-feng
New Design to Adaptive Neural Asymptotic Tracking Control for a Class of Uncertain Stochastic Nonlinear Systems With Unknown Input Constraints
In this paper, we present a new scheme to design an adaptive neural backstepping tracking controller for a class of stochastic nonlinear systems with unknown input constraints …
yiyun-liu
Fault Compensation-Based Adaptive Tracking Control for Nonlinear Systems with Actuator Saturation and Multiple Sensor Faults
An adaptive tracking control scheme based on a multi-dimensional Taylor network(MTN) is proposed for nonlinear systems with multiple sensor faults and actuator saturation. This …
yaoyao-guo
ResCapsnet: A Capsule Network with CRAM and BiGRU for Sound Event Detection
Sound event detection (SED) is a challenging task where ambient sound events are detected from a given audio signal, which includes categorizing the events and estimating their …
bing-sun
ASiT-CRNN: A Method for Sound Event Detection with Fine-Tuning of Self-Supervised Pre-Trained ASiT-based Model
Recently, the utilization of pre-trained models to transfer knowledge to downstream tasks has become an increasing trend. In this paper, we present an effective sound event …
yueyang-zheng
Sea Surface Temperature and Marine Heat Wave Predictions in the South China Sea: A 3D U-Net Deep Learning Model Integrating Multi-Source Data
Accurate sea surface temperature (SST) prediction is vital for disaster prevention, ocean circulation, and climate change. Traditional SST prediction methods, predominantly reliant …
bowen-xie
Event‐Triggered Adaptive Asymptotic Tracking Control for Stochastic Non‐Linear Systems With Unknown Hysteresis: A New Switching Threshold Approach
This paper proposes a novel event-triggered adaptive asymptotic tracking control (ATC) method for stochastic non-linear systems with unknown hysteresis. Firstly, in order to reduce …
yang-du
Warming Structure of the Kuroshio in the East China Sea since 1970
The Kuroshio Current (KC) has witnessed rapid surface warming during the past half-century, impacting the marine ecosystems in surrounding regions. However, the vertical structure …
xiaoqian-li