A5004972663- Risk and Portfolio Optimization
- Spectroscopy and Chemometric Analyses
- Spectroscopy Techniques in Biomedical and Chemical Research
- Probabilistic and Robust Engineering Design
- Advanced MIMO Systems Optimization
- Domain Adaptation and Few-Shot Learning
- Cooperative Communication and Network Coding
- Semantic Web and Ontologies
- Service-Oriented Architecture and Web Services
- Sparse and Compressive Sensing Techniques
- Advanced Wireless Communication Technologies
- Stochastic Gradient Optimization Techniques
- Generative Adversarial Networks and Image Synthesis
- Internet Traffic Analysis and Secure E-voting
- Model Reduction and Neural Networks
- Markov Chains and Monte Carlo Methods
- Integrated Energy Systems Optimization
- Advanced Chemical Sensor Technologies
- Advanced Statistical Process Monitoring
- Adversarial Robustness in Machine Learning
- Advanced Wireless Network Optimization
- Optical Wireless Communication Technologies
- Wireless Communication Security Techniques
- Statistical Methods and Inference
- Wireless Communication Networks Research
Tianjin Chengjian University
2025
Shenzhen Maternity and Child Healthcare Hospital
2025
Shantou University
2025
The University of Texas at Austin
2017-2024
Guangzhou University
2024
Nanjing University of Posts and Telecommunications
2020-2023
Xinjiang University
2020-2022
Fifth Tianjin Central Hospital
2022
University of Shanghai for Science and Technology
2021
Digital Hollywood University
2021
Distributionally robust stochastic optimization (DRSO) is an approach to under uncertainty in which, instead of assuming that there a known true underlying probability distribution, one hedges against chosen set distributions. In this paper, we first point out the distributions should be appropriate for application at hand and some choices have been popular until recently are, many applications, not good choices. We next consider sets are within Wasserstein distance from nominal...
Gas-fired units and power-to-gas facilities provide pivotal backups for power systems with volatile renewable generations. The deepened system interdependence calls elaborate consideration of network models both natural gas systems, as well uncertain factors. This paper proposes a data-driven distributionally robust optimization model the optimal gas-power flow problem wind generation. concept zonal line pack reserve are raised to topologically distinguish fuel suppliers gas-fired ensure...
This paper builds a bridge between two area in optimization and machine learning by establishing general connection Wasserstein distributional robustness variation regularization. It helps to demystify the empirical success of distributionally robust devise new regularization schemes for learning.
In this paper, we propose a risk-based data-driven distributionally robust approach to investigating the optimal power flow with dynamic line rating. The risk terms, including penalties for load shedding, wind generation curtailment and overload, are embedded into objective function. To robustify solution, consider distributional uncertainty set based on second-order moment, that captures correlation between outputs ratings, also Wasserstein distance, hedges against data perturbations. We...
Wasserstein distributionally robust optimization is a recent emerging modeling paradigm for decision making under data uncertainty. Because of its computational tractability and interpretability, it has achieved great empirical successes across several application domains in operations research, computer science, engineering, business analytics. Despite success, existing performance guarantees generic problems are not yet satisfactory. In this paper, we develop the first finite-sample...
We study policy optimization for the feature-based newsvendor, which seeks an end-to-end that renders explicit mapping from features to ordering decisions. Most existing works restrict policies some parametric class may suffer suboptimality (such as affine class) or lack of interpretability neural networks). Differently, we aim optimize over all functions features. In this case, classic empirical risk minimization yields a is not well-defined on unseen feature values. To avoid such...
Cardiovascular systems essentially have multiscale control mechanisms. Multiscale entropy (MSE) analysis permits the dynamic characterization of cardiovascular time series for both short-term and long-term processes, thus can be more illuminating. The traditional MSE heart rate variability (HRV) is performed on original RR interval (named as MSE_RR). In this study, we proposed an differential signal, named MSE_dRR. motivation using signal that has a direct link with inherent non-linear...
In this work, we investigate the relationship between reliability and security of a typical two-user downlink non-orthogonal multiple access (NOMA) communication system. The level successive interference cancellation on NOMA user is considered. Firstly, impact various key parameters transmit signal-to-noise ratio (SNR) users with outage probability (ROP) constraint discussed. Taking minimum SNR for ROP into account, secrecy performance systems studied analytical expressions system are...
Distributionally robust stochastic optimization (DRSO) is a framework for decision-making problems under certainty, which finds solutions that perform well chosen set of probability distributions. Many different approaches specifying distributions have been proposed. The choice matters, because it affects the results, and relative performance choices depend on characteristics problems. In this paper, we consider in random variables exhibit some form dependence, but exact values parameters...
Zero energy building (ZEB) is one of the important ways to reduce carbon emissions. Therefore, this paper proposed a new off-grid integrated system for ZEB (ZEBOIES) and its optimal scheduling strategy based on twin delayed deep deterministic policy gradient with mixed integer linear programming constrained (MC-TD3) algorithm. First, topological structure ZEBOIES described in detail, then mathematical model each component described. Second, bilateral uncertainty supply demand ZEBOIES,...
In existing breast cancer prediction research, most models rely solely on a single type of imaging data, which limits their performance. To overcome this limitation, the present study explores based multimodal medical images (mammography and ultrasound images) compares them with single-modal models. We collected data from 790 patients, including 2,235 mammography 1,348 images, conducted comparison using six deep learning classification to identify best model for constructing model....
As a characteristic edible fungus with high nutritional value and medicinal effect, the Bachu mushroom has broad market. To distinguish among mushrooms other fungi effectively accurately, as well to explore universal identification method, this study proposed method identify by Fourier Transform Infrared Spectroscopy (FT-IR) combined machine learning. In experiment, two kinds of common mushrooms, Lentinus edodes club fungi, were selected classified mushrooms. Due different distribution...
Abstract To improve the efficiency and quality of college English teaching, we analyzed feasibility application process data mining technology in teaching. The entire classification was fully realized. A new teaching program proposed. object target were determined. Online surveys used to collect data. Data integration, cleaning, conversion, reduction other pre-processing technologies adopted. decision tree generated by using C4.5 algorithm, pruning carried out. result analysis model...
We develop a novel computationally efficient and general framework for robust hypothesis testing. The new features way to construct uncertainty sets under the null alternative distributions, which are centered around empirical distribution defined via Wasserstein metric, thus our approach is data-driven free of distributional assumptions. convex safe approximation minimax formulation show that such renders nearly-optimal detector among family all possible tests. By exploiting structure least...