arXiv subjects

Nan Yang

Publications and source records attributed to Nan Yang.

At least 127 records · Page 7Linked to original sources

arXiv2023

FedMAE: Federated Self-Supervised Learning with One-Block Masked Auto-Encoder

Latest federated learning (FL) methods started to focus on how to use unlabeled data in clients for training due to users' privacy concerns, high labeling costs, or lack of expertise. However, current Federated Semi-Supervised/Self-Supervised Learning (FSSL) approaches fail to learn large-scale images because of the limited computing resources of local clients. In this paper, we introduce a new framework FedMAE, which stands for Federated Masked AutoEncoder, to address the problem of how to utilize unlabeled large-scale images for FL. Specifically, FedMAE can pre-train one-block Masked AutoEncoder (MAE) using large images in lightweight client devices, and then cascades multiple pre-trained one-block MAEs in the server to build a multi-block ViT backbone for downstream tasks. Theoretical analysis and experimental results on image reconstruction and classification show that our FedMAE achieves superior performance compared to the state-of-the-art FSSL methods.

cs.LG↗

arXiv2023

Real-time scheduling of renewable power systems through planning-based reinforcement learning

The growing renewable energy sources have posed significant challenges to traditional power scheduling. It is difficult for operators to obtain accurate day-ahead forecasts of renewable generation, thereby requiring the future scheduling system to make real-time scheduling decisions aligning with ultra-short-term forecasts. Restricted by the computation speed, traditional optimization-based methods can not solve this problem. Recent developments in reinforcement learning (RL) have demonstrated the potential to solve this challenge. However, the existing RL methods are inadequate in terms of constraint complexity, algorithm performance, and environment fidelity. We are the first to propose a systematic solution based on the state-of-the-art reinforcement learning algorithm and the real power grid environment. The proposed approach enables planning and finer time resolution adjustments of power generators, including unit commitment and economic dispatch, thus increasing the grid's ability to admit more renewable energy. The well-trained scheduling agent significantly reduces renewable curtailment and load shedding, which are issues arising from traditional scheduling's reliance on inaccurate day-ahead forecasts. High-frequency control decisions exploit the existing units' flexibility, reducing the power grid's dependence on hardware transformations and saving investment and operating costs, as demonstrated in experimental results. This research exhibits the potential of reinforcement learning in promoting low-carbon and intelligent power systems and represents a solid step toward sustainable electricity generation.

cs.AI↗

arXiv2023

FedIL: Federated Incremental Learning from Decentralized Unlabeled Data with Convergence Analysis

Most existing federated learning methods assume that clients have fully labeled data to train on, while in reality, it is hard for the clients to get task-specific labels due to users' privacy concerns, high labeling costs, or lack of expertise. This work considers the server with a small labeled dataset and intends to use unlabeled data in multiple clients for semi-supervised learning. We propose a new framework with a generalized model, Federated Incremental Learning (FedIL), to address the problem of how to utilize labeled data in the server and unlabeled data in clients separately in the scenario of Federated Learning (FL). FedIL uses the Iterative Similarity Fusion to enforce the server-client consistency on the predictions of unlabeled data and uses incremental confidence to establish a credible pseudo-label set in each client. We show that FedIL will accelerate model convergence by Cosine Similarity with normalization, proved by Banach Fixed Point Theorem. The code is available at https://anonymous.4open.science/r/fedil.

cs.LG↗

arXiv2023

Random Padding Data Augmentation

The convolutional neural network (CNN) learns the same object in different positions in images, which can improve the recognition accuracy of the model. An implication of this is that CNN may know where the object is. The usefulness of the features' spatial information in CNNs has not been well investigated. In this paper, we found that the model's learning of features' position information hindered the learning of the features' relationship. Therefore, we introduced Random Padding, a new type of padding method for training CNNs that impairs the architecture's capacity to learn position information by adding zero-padding randomly to half of the border of feature maps. Random Padding is parameter-free, simple to construct, and compatible with the majority of CNN-based recognition models. This technique is also complementary to data augmentations such as random cropping, rotation, flipping and erasing, and consistently improves the performance of image classification over strong baselines.

cs.CV↗

arXiv2022

Title2Event: Benchmarking Open Event Extraction with a Large-scale Chinese Title Dataset

Event extraction (EE) is crucial to downstream tasks such as new aggregation and event knowledge graph construction. Most existing EE datasets manually define fixed event types and design specific schema for each of them, failing to cover diverse events emerging from the online text. Moreover, news titles, an important source of event mentions, have not gained enough attention in current EE research. In this paper, We present Title2Event, a large-scale sentence-level dataset benchmarking Open Event Extraction without restricting event types. Title2Event contains more than 42,000 news titles in 34 topics collected from Chinese web pages. To the best of our knowledge, it is currently the largest manually-annotated Chinese dataset for open event extraction. We further conduct experiments on Title2Event with different models and show that the characteristics of titles make it challenging for event extraction, addressing the significance of advanced study on this problem. The dataset and baseline codes are available at https://open-event-hub.github.io/title2event.

cs.CL↗

arXiv2022

Age of Information of Multi-user Mobile Edge Computing Systems

In this paper, we analyze the average age of information (AoI) and the average peak AoI (PAoI) of a multiuser mobile edge computing (MEC) system where a base station (BS) generates and transmits computation-intensive packets to user equipments (UEs). In this MEC system, we focus on three computing schemes: (i) The local computing scheme where all computational tasks are computed by the local server at the UE, (ii) The edge computing scheme where all computational tasks are computed by the edge server at the BS, and (iii) The partial computing scheme where computational tasks are partially allocated at the edge server and the rest are computed by the local server. Considering exponentially distributed transmission time and computation time and adopting the first come first serve (FCFS) queuing policy, we derive closed-form expressions for the average AoI and average PAoI. To address the complexity of the average AoI expression, we derive simple upper and lower bounds on the average AoI, which allow us to explicitly examine the dependence of the optimal offloading decision on the MEC system parameters. Aided by simulation results, we verify our analysis and illustrate the impact of system parameters on the AoI performance.

cs.IT↗

arXiv2022

Average Age of Information Penalty of Short-Packet Communications with Packet Management

In this paper, we analyze the non-linear age of information (AoI) performance in a point-to-point short packet communication system, where a transmitter generates packets based on status updates and transmits the packets to a receiver. Specifically, we investigate three packet management strategies, namely, the non-preemption with no buffer strategy, the non-preemption with one buffer strategy, and the preemption strategy. To characterize the level of the receiver's dissatisfaction on outdated data, we adopt a generalized α-βAoI penalty function into the analysis and derive closed-form expressions for the average AoI penalty achieved by the three packet management strategies. Simulation results are used to corroborate our analysis and explicitly evaluate the impact of various system parameters, such as the coding rate and status update generation rate, on the AoI performance. Additionally, we find that the value of αreflects the system transmission reliability.

cs.IT↗

arXiv2022

Terahertz Communications for Massive Connectivity and Security in 6G and Beyond Era

Terahertz (THz) communications (THzCom) has experienced a meteoric rise of interest, due to its benefits for ultra-high data rate transmission in the sixth generation (6G) and beyond era. Despite so, the research on exploring the potential of THzCom for other performance targets anticipated by 6G, including massive connectivity and security, is still in its infancy. In this article, we start with briefly describing the unique peculiarities of THz channels, and then discuss theoretical frameworks to facilitate the analysis and design of THz transmission for achieving massive connectivity and security. Then we discuss promising spectrum management strategies, including the exploration of multiple THz transmission windows and frequency reuse with multiplexing and signal processing, to substantially increase the number of supported users and identify to-be-tackled challenges. We further present important research directions based on the principles of physical layer security, such as new spectrum allocation policies and beamforming algorithms, to fight against eavesdropping in THzCom systems, ushering in secure THzCom systems.

eess.SP↗

arXiv2022

Adversarial Transformer for Repairing Human Airway Segmentation

Discontinuity in the delineation of peripheral bronchioles hinders the potential clinical application of automated airway segmentation models. Moreover, the deployment of such models is limited by the data heterogeneity across different centres, and pathological abnormalities also make achieving accurate robust segmentation in distal small airways difficult. Meanwhile, the diagnosis and prognosis of lung diseases often rely on evaluating structural changes in those anatomical regions. To address this gap, this paper presents a patch-scale adversarial-based refinement network that takes in preliminary segmentation along with original CT images and outputs a refined mask of the airway structure. The method is validated on three different datasets encompassing healthy cases, cases with cystic fibrosis and cases with COVID-19. The results are quantitatively evaluated by seven metrics and achieved more than a 15% rise in detected length ratio and detected branch ratio, showing promising performance compared to previously proposed models. The visual illustration also proves our refinement guided by a patch-scale discriminator and centreline objective functions is effective in detecting discontinuities and missing bronchioles. Furthermore, the generalizability of our refinement pipeline is tested on three previous models and improves their segmentation completeness significantly.

eess.IV↗

arXiv2022

CCR: Facial Image Editing with Continuity, Consistency and Reversibility

Three problems exist in sequential facial image editing: incontinuous editing, inconsistent editing, and irreversible editing. Incontinuous editing is that the current editing can not retain the previously edited attributes. Inconsistent editing is that swapping the attribute editing orders can not yield the same results. Irreversible editing means that operating on a facial image is irreversible, especially in sequential facial image editing. In this work, we put forward three concepts and corresponding definitions: editing continuity, consistency, and reversibility. Then, we propose a novel model to achieve the goal of editing continuity, consistency, and reversibility. A sufficient criterion is defined to determine whether a model is continuous, consistent, and reversible. Extensive qualitative and quantitative experimental results validate our proposed model and show that a continuous, consistent and reversible editing model has a more flexible editing function while preserving facial identity. Furthermore, we think that our proposed definitions and model will have wide and promising applications in multimedia processing. Code and data are available at https://github.com/mickoluan/CCR.

cs.CV↗

arXiv2022

Learning Diverse Document Representations with Deep Query Interactions for Dense Retrieval

In this paper, we propose a new dense retrieval model which learns diverse document representations with deep query interactions. Our model encodes each document with a set of generated pseudo-queries to get query-informed, multi-view document representations. It not only enjoys high inference efficiency like the vanilla dual-encoder models, but also enables deep query-document interactions in document encoding and provides multi-faceted representations to better match different queries. Experiments on several benchmarks demonstrate the effectiveness of the proposed method, out-performing strong dual encoder baselines.The code is available at \url{https://github.com/jordane95/dual-cross-encoder

cs.IR↗

arXiv2022

An Unsupervised Learning Approach for Spectrum Allocation in Terahertz Communication Systems

We propose a new spectrum allocation strategy, aided by unsupervised learning, for multiuser terahertz communication systems. In this strategy, adaptive sub-band bandwidth is considered such that the spectrum of interest can be divided into sub-bands with unequal bandwidths. This strategy reduces the variation in molecular absorption loss among the users, leading to the improved data rate performance. We first formulate an optimization problem to determine the optimal sub-band bandwidth and transmit power, and then propose the unsupervised learning-based approach to obtaining the near-optimal solution to this problem. In the proposed approach, we first train a deep neural network (DNN) while utilizing a loss function that is inspired by the Lagrangian of the formulated problem. Then using the trained DNN, we approximate the near-optimal solutions. Numerical results demonstrate that comparing to existing approaches, our proposed unsupervised learning-based approach achieves a higher data rate, especially when the molecular absorption coefficient within the spectrum of interest varies in a highly non-linear manner.

cs.LG↗

arXiv2022

Terahertz Communications for 6G and Beyond Wireless Networks: Challenges, Key Advancements, and Opportunities

The unprecedented increase in wireless data traffic, predicted to occur within the next decade, is motivating academia and industries to look beyond contemporary wireless standards and conceptualize the sixth-generation (6G) wireless networks. Among various promising solutions, terahertz (THz) communications (THzCom) is recognized as a highly promising technology for the 6G and beyond era, due to its unique potential to support terabit-per-second transmission in emerging applications. This article delves into key areas for developing end-to-end THzCom systems, focusing on physical, link, and network layers. Specifically, we discuss the areas of THz spectrum management, THz antennas and beamforming, and the integration of other 6G-enabling technologies for THzCom. For each area, we identify the challenges imposed by the unique properties of the THz band. We then present main advancements and outline perspective research directions in each area to stimulate future research efforts for realizing THzCom in 6G and beyond wireless networks.

eess.SP↗

arXiv2022

Novel Spectrum Allocation Among Multiple Transmission Windows for Terahertz Communication Systems

This paper presents a novel spectrum allocation strategy for multiuser terahertz (THz) band communication systems when the to-be-allocated spectrum is composed of multiple transmission windows (TWs). This strategy explores the benefits of (i) allowing users to occupy sub-bands with unequal bandwidths and (ii) optimally avoiding using some spectra that exist at the edges of TWs where molecular absorption loss is high. To maximize the aggregated multiuser data rate, we formulate an optimization problem, with the primary focus on spectrum allocation. We then apply transformations and modifications to make the problem computationally tractable, and develop an iterative algorithm based on successive convex approximation to determine the optimal sub-band bandwidth and the unused spectra at the edges of TWs. Using numerical results, we show that a significantly higher data rate can be achieved by changing the sub-band bandwidth, as compared to equal sub-band bandwidth. We also show that a further data rate gain can be obtained by optimally determining the unused spectra at the edges of TWs, as compared to avoiding using pre-defined spectra at the edges of TWs.

eess.SP↗

arXiv2022

Spectrum Allocation with Adaptive Sub-band Bandwidth for Terahertz Communication Systems

We study spectrum allocation for terahertz (THz) band communication (THzCom) systems, while considering the frequency and distance-dependent nature of THz channels. Different from existing studies, we explore multi-band-based spectrum allocation with adaptive sub-band bandwidth (ASB) by allowing the spectrum of interest to be divided into sub-bands with unequal bandwidths. Also, we investigate the impact of sub-band assignment on multi-connectivity (MC) enabled THzCom systems, where users associate and communicate with multiple access points simultaneously. We formulate resource allocation problems, with the primary focus on spectrum allocation, to determine sub-band assignment, sub-band bandwidth, and optimal transmit power. Thereafter, we propose reasonable approximations and transformations, and develop iterative algorithms based on the successive convex approximation technique to analytically solve the formulated problems. Aided by numerical results, we show that by enabling and optimizing ASB, significantly higher throughput can be achieved as compared to adopting equal sub-band bandwidth, and this throughput gain is most profound when the power budget constraint is more stringent. We also show that our sub-band assignment strategy in MC-enabled THzCom systems outperforms the state-of-the-art sub-band assignment strategies and the performance gain is most profound when the spectrum with the lowest average molecular absorption coefficient is selected during spectrum allocation.

eess.SP↗

arXiv2022

Molecular Absorption Effect: A Double-edged Sword of Terahertz Communications

Communications in the terahertz band (THz) (0.1--10~THz) have been regarded as a promising technology for future 6G and beyond wireless systems, to overcome the challenges of evergrowing wireless data traffic and crowded spectrum. As the frequency increases from the microwave band to the THz band, new spectrum features pose unprecedented challenges to wireless communication system design. The molecular absorption effect is one of the new THz spectrum properties, which enlarges the path loss and noise at specific frequencies. This brings in a double-edged sword for THz wireless communication systems. On one hand, from the data rate viewpoint, molecular absorption is detrimental, since it mitigates the received signal power and degrades the channel capacity. On the other hand, it is worth noticing that for wireless security and covertness, the molecular absorption effect can be utilized to safeguard THz communications among users. In this paper, the features of the molecular absorption effect and their impact on the THz system design are analyzed under various scenarios, with the ultimate goal of providing guidelines to how better exploit this unique THz phenomenon. Specifically, since the molecular absorption greatly depends on the propagation medium, different communication scenarios consisting of various media are discussed, including terrestrial, air and space, sea surface and nano-scale communications. Furthermore, two novel molecular absorption enlightened secure and covert communication schemes are presented, where the molecular absorption effect is utilized as the key and unique feature to boost security and covertness.

cs.IT↗

arXiv2022

Analysis of MC Systems Employing Receivers Covered by Heterogeneous Receptors

This paper investigates the channel impulse response (CIR), i.e., the molecule hitting rate, of a molecular communication (MC) system employing an absorbing receiver (RX) covered by multiple non overlapping receptors. In this system, receptors are heterogeneous, i.e., they may have different sizes and arbitrary locations. Furthermore, we consider two types of transmitter (TX), namely a point TX and a membrane fusion (MF)-based spherical TX. We assume the point TX or the center of the MF-based TX has a fixed distance to the center of the RX. Given this fixed distance, the TX can be at different locations and the CIR of the RX depends on the exact location of the TX. By averaging over all possible TX locations, we analyze the expected molecule hitting rate at the RX as a function of the sizes and locations of the receptors, where we assume molecule degradation may occur during the propagation of the signaling molecules. Notably, our analysis is valid for different numbers, a wide range of sizes, and arbitrary locations of the receptors, and its accuracy is confirmed via particle-based simulations. Exploiting our numerical results, we show that the expected number of absorbed molecules at the RX increases with the number of receptors, when the total area on the RX surface covered by receptors is fixed. Based on the derived analytical expressions, we compare different geometric receptor distributions by examining the expected number of absorbed molecules at the RX. We show that evenly distributed receptors result in a larger number of absorbed molecules than other distributions. We further compare three models that combine different types of TXs and RXs.

cs.ET↗

arXiv2022

Fast Parallel Hypertree Decompositions in Logarithmic Recursion Depth

Modern trends in data collection are bringing current mainstream techniques for database query processing to their limits. Consequently, various novel approaches for efficient query processing are being actively studied. One such approach is based on hypertree decompositions (HDs), which have been shown to carry great potential to process complex queries more efficiently and with stronger theoretical guarantees. However, using HDs for query execution relies on the difficult task of computing decompositions of the query structure, which guides the efficient execution of the query. From theoretical results we know that the performance of purely sequential methods is inherently limited, yet the problem is susceptible to parallelisation. In this paper we propose the first algorithm for computing hypertree decompositions that is well-suited for parallelisation. The proposed algorithm log-k-decomp requires only a logarithmic number of recursion levels and additionally allows for highly parallelised pruning of the search space by restriction to balanced separators. We provide detailed experimental evaluation over the HyperBench benchmark and demonstrate that our approach is highly effective especially for complex queries.

cs.DB↗