Searcharxiv⌕ Search

arXiv subjects

Hailong Huang

Publications and source records attributed to Hailong Huang.

24 records · Page 2Linked to original sources
arXiv2022

Machine Learning-Aided Operations and Communications of Unmanned Aerial Vehicles: A Contemporary Survey

The ongoing amalgamation of UAV and ML techniques is creating a significant synergy and empowering UAVs with unprecedented intelligence and autonomy. This survey aims to provide a timely and comprehensive overview of ML techniques used in UAV operations and communications and identify the potential growth areas and research gaps. We emphasise the four key components of UAV operations and communications to which ML can significantly contribute, namely, perception and feature extraction, feature interpretation and regeneration, trajectory and mission planning, and aerodynamic control and operation. We classify the latest popular ML tools based on their applications to the four components and conduct gap analyses. This survey also takes a step forward by pointing out significant challenges in the upcoming realm of ML-aided automated UAV operations and communications. It is revealed that different ML techniques dominate the applications to the four key modules of UAV operations and communications. While there is an increasing trend of cross-module designs, little effort has been devoted to an end-to-end ML framework, from perception and feature extraction to aerodynamic control and operation. It is also unveiled that the reliability and trust of ML in UAV operations and applications require significant attention before full automation of UAVs and potential cooperation between UAVs and humans come to fruition.

cs.RO↗
arXiv2021

A Diversity-Enhanced and Constraints-Relaxed Augmentation for Low-Resource Classification

Data augmentation (DA) aims to generate constrained and diversified data to improve classifiers in Low-Resource Classification (LRC). Previous studies mostly use a fine-tuned Language Model (LM) to strengthen the constraints but ignore the fact that the potential of diversity could improve the effectiveness of generated data. In LRC, strong constraints but weak diversity in DA result in the poor generalization ability of classifiers. To address this dilemma, we propose a {D}iversity-{E}nhanced and {C}onstraints-\{R}elaxed {A}ugmentation (DECRA). Our DECRA has two essential components on top of a transformer-based backbone model. 1) A k-beta augmentation, an essential component of DECRA, is proposed to enhance the diversity in generating constrained data. It expands the changing scope and improves the degree of complexity of the generated data. 2) A masked language model loss, instead of fine-tuning, is used as a regularization. It relaxes constraints so that the classifier can be trained with more scattered generated data. The combination of these two components generates data that can reach or approach category boundaries and hence help the classifier generalize better. We evaluate our DECRA on three public benchmark datasets under low-resource settings. Extensive experiments demonstrate that our DECRA outperforms state-of-the-art approaches by 3.8% in the overall score.

cs.CL↗
arXiv2021

Adversarial Mixing Policy for Relaxing Locally Linear Constraints in Mixup

Mixup is a recent regularizer for current deep classification networks. Through training a neural network on convex combinations of pairs of examples and their labels, it imposes locally linear constraints on the model's input space. However, such strict linear constraints often lead to under-fitting which degrades the effects of regularization. Noticeably, this issue is getting more serious when the resource is extremely limited. To address these issues, we propose the Adversarial Mixing Policy (AMP), organized in a min-max-rand formulation, to relax the Locally Linear Constraints in Mixup. Specifically, AMP adds a small adversarial perturbation to the mixing coefficients rather than the examples. Thus, slight non-linearity is injected in-between the synthetic examples and synthetic labels. By training on these data, the deep networks are further regularized, and thus achieve a lower predictive error rate. Experiments on five text classification benchmarks and five backbone models have empirically shown that our methods reduce the error rate over Mixup variants in a significant margin (up to 31.3%), especially in low-resource conditions (up to 17.5%).

cs.CL↗
arXiv2017

Optimized Deployment of Autonomous Drones to Improve User Experience in Cellular Networks

Modern wireless traffic demand pushes Internet Service Providers to develop effective strategies to improve user experience. Since deploying dense Base Stations (BSs) is not cost efficient, an alternative is to deploy autonomous drones to supplement existing BSs. A street graph is adopted to represent the area of interest. The outdoor User Equipments (UEs) to be served locate near streets and the 2D projections of drones are restricted to streets to avoid collision with buildings. We build up a UE density function based on a real dataset, reflecting the traffic in the area. We study four problems: where to deploy single drone to cover maximum UEs, where to deploy $k$ drones cover maximum UEs subject to an inner drone distance constraint, where to deploy $k$ drones cover maximum UEs subject to inner drone distance constraint and drones' battery constraints, and the minimum drones to cover a given percentage of UEs subject to inner drone distance constraint. We prove that the latter three problems are NP-hard and propose greedy algorithms with theoretical analysis. To our best knowledge, this is the first paper to consider the battery constraints for drone deployments. Extensive simulations have been conducted to verify the effectiveness of our approaches.

cs.NI↗
arXiv2017

Performance Improvement by Introducing Mobility in Wireless Communication Networks

Communication technology is a major contributor to our lifestyles. Improving the performance of communication system brings with various benefits to human beings. This report covers two typical such systems: wireless sensor networks and cellular networks. We study the influence of mobility in these networks. In wireless sensor networks, we consider the usage of mobile sinks to collect sensory data from static nodes. For the latter case, we design data collection protocols for a single mobile sink which aim at balancing the energy consumption among sensor nodes to improve the network lifetime. We also design an algorithm for multiple mobile sinks to collect urgent sensory information within the allowed latency. Regarding the mobility in cellular networks, we mean the service providers are mobile. Conventional approaches usually consider how to optimally deploy static base stations according to a certain metric, such as throughput. Base station densification is a solution, but it is not cost efficient. Introducing mobility is to let the base stations fly in the sky, resulting in less investment. We consider one of the key issues of using drones to serve cellular users: drone deployment. We have implemented all the proposed algorithms by either simulations or experiments, and the results have confirmed the effectiveness of these approaches.

cs.NI↗