A5016833874- Distributed Control Multi-Agent Systems
- Sparse and Compressive Sensing Techniques
- Cooperative Communication and Network Coding
- Neural Networks Stability and Synchronization
- Stochastic Gradient Optimization Techniques
- Mathematical and Theoretical Epidemiology and Ecology Models
- Sex work and related issues
- Electric Power System Optimization
- Digital Rights Management and Security
- Smart Grid Security and Resilience
- Smart Grid Energy Management
- Anomaly Detection Techniques and Applications
- HIV, Drug Use, Sexual Risk
- UAV Applications and Optimization
- Energy Load and Power Forecasting
- Electricity Theft Detection Techniques
- Advanced Memory and Neural Computing
- Image and Signal Denoising Methods
- Advanced Computational Techniques and Applications
- HIV/AIDS Research and Interventions
Nanyang Technological University
2022-2023
Southwest University
2020-2023
Jiangxi University of Traditional Chinese Medicine
2022
Institute of Economics
2022
Nanjing University of Chinese Medicine
2017-2018
Over the past decade, domain of deep learning has emerged as a swiftly expanding area interest among researchers globally. It offers substantial benefits over traditional shallow networks, particularly in realms feature extraction and model fitting. Deep excels at uncovering intricate, distributed features from raw data, which exhibit robust generalization capabilities. triumphantly addressed challenges that were once deemed intractable within field artificial intelligence. With exponential...
Decentralized dual methods play significant roles in large-scale optimization, which effectively resolve many constrained optimization problems machine learning and power systems. In this article, we focus on studying a class of totally non-smooth composite over multi-agent systems, where the mutual goal agents system is to optimize sum two separable functions consisting strongly-convex function another convex (not necessarily strongly-convex) function. Agents conduct parallel local...
To detect cyber-attacks on individual smart meters, this letter proposes a novel data-driven method based dimensional augmentation with recurrence plots (RPs) and visual geometry group network (VGGNet). Firstly, the real-time 1-dimensional time-series meter data is augmented to 2-dimensional image using RPs method, which provides visual-distinguishable features that can be more easily identified by computer vision-based algorithms. Then, genuine contaminated are distinguished VGGNet data....
Due to the stochastic nature of occupants' behaviors, forecasting individual household-level residential load has been a challenging problem. In this paper, by correlating habitual electricity usages and short-term variances, new prediction intervals (PIs) based method combining deep learning error ranges estimation is proposed. Firstly, double-layer Long Short-Time Memory (LSTM) model constructed predict hourly demand at household-level. The statistical errors analysis shows that LSTM have...
Abstract This paper studies distributed convex optimization problems over an undirected network where all nodes cooperate to minimize a sum of local objective functions. Each function is further assumed be average several instantaneous By incorporating the stochastic averaging gradient into first‐order primal‐dual method, algorithm with multi‐step communication proposed solve problem. For each node, one randomly selected evaluated per iteration, which effectively reduces computation cost...
The ongoing spread of HIV after sobering news about the goal “End AIDS” is not encouraging, apart from regional differences. We focus on consequences two essentially failed prevention strategies in certain countries. first because correct messages concerning preventive behavior did reach required levels target populations to interrupt infection chains. There was a lack appropriate framework conditions for engage scale. additional biomedical strategy “Treatment as Prevention” didn’t achieve...
This paper investigates a distributed optimization problem over multiagent network, in which the target of agents is to collaboratively optimize sum all local objective functions. The case discusses that network topology among described by strongly connected directed graph. proposed algorithm utilizes row-stochastic weight matrices and uncoordinated step sizes. Under conditions functions are strong convex, have Lipschitz continuous gradients, we manifest faster linearly converges global...
This paper proposes an edge-based distributed primal-dual algorithm for economic dispatch problem (EDP), in which a collection of networked generators cooperatively minimizes the sum all local cost functions while satisfying some coupling constraints. Different from existing results, this applies techniques, such as factorization weighted Laplacian and spectral decomposition, to prove convergence proposed algorithm. It proves that converges provided step-size is positive lower than exact...