Guo

Ziyue Li, Qianwen, Guo et al.

Drone delivery is rapidly emerging as a cost-effective and energy efficient alternative for last-mile delivery. Unlike ground vehicles, a drone's energy consumption depends on its payload in addition to travel distance. This creates a unique environmental challenge for multi-stop delivery tours, as the drone's total weight, and therefore its energy consumption rate, dynamically changes after each delivery. This paper investigates a novel green drone routing problem focused on maximizing energy efficiency. Through a series of motivating examples and numerical experiments, we demonstrate that energy-aware routing leads to several counter-intuitive routing strategies that contradict traditional distance-minimization delivery: a longer route may actually consume less energy than a shorter one; separate single-customer tours can be superior to a multi-stop tour; and a heterogeneous fleet, with drones of varying sizes, can achieve greater efficiency by matching drone capacity to specific delivery demands. In the numerical study, the green routing strategy shows energy savings in 67% of the instances. For these cases, the average energy saving is 2.17%, with a maximum saving of 5.97%, compared to minimum distance routing. These findings highlight the potential for green drone routing strategies to improve the sustainability of last-mile delivery.

Yinan Wang, Diane Oyen, Weihong et al.

Catastrophic failure in brittle materials is often due to the rapid growth and coalescence of cracks aided by high internal stresses. Hence, accurate prediction of maximum internal stress is critical to predicting time to failure and improving the fracture resistance and reliability of materials. Existing high-fidelity methods, such as the Finite-Discrete Element Model (FDEM), are limited by their high computational cost. Therefore, to reduce computational cost while preserving accuracy, a novel deep learning model, "StressNet," is proposed to predict the entire sequence of maximum internal stress based on fracture propagation and the initial stress data. More specifically, the Temporal Independent Convolutional Neural Network (TI-CNN) is designed to capture the spatial features of fractures like fracture path and spall regions, and the Bidirectional Long Short-term Memory (Bi-LSTM) Network is adapted to capture the temporal features. By fusing these features, the evolution in time of the maximum internal stress can be accurately predicted. Moreover, an adaptive loss function is designed by dynamically integrating the Mean Squared Error (MSE) and the Mean Absolute Percentage Error (MAPE), to reflect the fluctuations in maximum internal stress. After training, the proposed model is able to compute accurate multi-step predictions of maximum internal stress in approximately 20 seconds, as compared to the FDEM run time of 4 hours, with an average MAPE of 2% relative to test data.

Jiyang Gao, Zijian, Guo et al.

Fine-grained image labels are desirable for many computer vision applications, such as visual search or mobile AI assistant. These applications rely on image classification models that can produce hundreds of thousands (e.g. 100K) of diversified fine-grained image labels on input images. However, training a network at this vocabulary scale is challenging, and suffers from intolerable large model size and slow training speed, which leads to unsatisfying classification performance. A straightforward solution would be training separate expert networks (specialists), with each specialist focusing on learning one specific vertical (e.g. cars, birds...). However, deploying dozens of expert networks in a practical system would significantly increase system complexity and inference latency, and consumes large amounts of computational resources. To address these challenges, we propose a Knowledge Concentration method, which effectively transfers the knowledge from dozens of specialists (multiple teacher networks) into one single model (one student network) to classify 100K object categories. There are three salient aspects in our method: (1) a multi-teacher single-student knowledge distillation framework; (2) a self-paced learning mechanism to allow the student to learn from different teachers at various paces; (3) structurally connected layers to expand the student network capacity with limited extra parameters. We validate our method on OpenImage and a newly collected dataset, Entity-Foto-Tree (EFT), with 100K categories, and show that the proposed model performs significantly better than the baseline generalist model.