arXiv subjects

Weihong "Grace" Guo

Publications and source records attributed to Weihong "Grace" Guo.

2 recordsLinked to original sources

arXiv2024

A simulation-optimization framework for food supply chain network design to ensure food accessibility under uncertainty

How to ensure accessibility to food and nutrition while food supply chains suffer from demand and supply uncertainties caused by disruptive forces such as the COVID-19 pandemic and natural disasters is an emerging and critical issue. Unstable access to food influences the level of nutrition that weakens the health and well-being of citizens. Therefore, a food accessibility evaluation index is proposed in this work to quantify how well nutrition needs are met. The proposed index is then embedded in a stochastic multi-objective mixed-integer optimization problem to determine the optimal supply chain design to maximize food accessibility and minimize cost. Considering uncertainty in demand and supply, the multi-objective problem is solved in a two-phase simulation-optimization framework in which Green Field Analysis is applied to determine the long-term, tactical decisions such as supply chain configuration, and then Monte Carlo simulation is performed iteratively to determine the short-term supply chain operations by solving a stochastic programming problem. A case study is conducted on the beef supply chain in Qatar. Pareto efficient solutions are validated in discrete event simulation to evaluate the performance of the designed supply chain in various realistic scenarios and provide recommendations for different decision-makers.

econ.GN↗

arXiv2021

Tensor decomposition to Compress Convolutional Layers in Deep Learning

Feature extraction for tensor data serves as an important step in many tasks such as anomaly detection, process monitoring, image classification, and quality control. Although many methods have been proposed for tensor feature extraction, there are still two challenges that need to be addressed: 1) how to reduce the computation cost for high dimensional and large volume tensor data; 2) how to interpret the output features and evaluate their significance. {The most recent methods in deep learning, such as Convolutional Neural Network (CNN), have shown outstanding performance in analyzing tensor data, but their wide adoption is still hindered by model complexity and lack of interpretability. To fill this research gap, we propose to use CP-decomposition to approximately compress the convolutional layer (CPAC-Conv layer) in deep learning. The contributions of our work could be summarized into three aspects: (1) we adapt CP-decomposition to compress convolutional kernels and derive the expressions of both forward and backward propagations for our proposed CPAC-Conv layer; (2) compared with the original convolutional layer, the proposed CPAC-Conv layer can reduce the number of parameters without decaying prediction performance. It can combine with other layers to build novel deep Neural Networks; (3) the value of decomposed kernels indicates the significance of the corresponding feature map, which provides us with insights to guide feature selection.

cs.LG↗