Search

The Agriculture-Vision Challenge at CVPR 2024 aims at leveraging semantic segmentation models to produce pixel level semantic segmentation labels within regions of interest for multi-modality satellite images. It is one of the most famous and competitive challenges for global researchers to break the boundary between computer vision and agriculture sectors. However, there is a serious class imbalance problem in the agriculture-vision dataset, which hinders the semantic segmentation performance. To solve this problem, firstly, we propose a mosaic data augmentation with a rare class sampling strategy to enrich long-tail class samples. Secondly, we employ an adaptive class weight scheme to suppress the contribution of the common classes while increasing the ones of rare classes. Thirdly, we propose a probability post-process to increase the predicted value of the rare classes. Our methodology achieved a mean Intersection over Union (mIoU) score of 0.547 on the test set, securing second place in this challenge.

The spread of a graph $G$ is the difference between the largest and smallest eigenvalue of the adjacency matrix of $G$. In this paper, we consider the family of graphs which contain no $K_{s,t}$-minor. We show that for any $t\geq s \geq 2$, there is an integer $\xi_{t}$ such that the extremal $n$-vertex $K_{s,t}$-minor-free graph attaining the maximum spread is the graph obtained by joining a graph $L$ on $(s-1)$ vertices to the disjoint union of $\lfloor \frac{2n+\xi_{t}}{3t}\rfloor$ copies of $K_t$ and $n-s+1 - t\lfloor \frac{2n+\xi_t}{3t}\rfloor$ isolated vertices. Furthermore, we give an explicit formula for $\xi_{t}$ and an explicit description for the graph $L$ for $t \geq \frac32(s-3) +\frac{4}{s-1}$., Comment: 21 pages. arXiv admin note: text overlap with arXiv:2212.05540

For integers $k>\ell\ge0$, a graph $G$ is $(k,\ell)$-stable if $\alpha(G-S)\geq \alpha(G)-\ell$ for every $S\subseteq V(G)$ with $|S|=k$. A recent result of Dong and Wu [SIAM J. Discrete Math., 36 (2022) 229--240] shows that every $(k,\ell)$-stable graph $G$ satisfies $\alpha(G) \le \lfloor ({|V(G)|-k+1})/{2}\rfloor+\ell$. A $(k,\ell)$-stable graph $G$ is tight if $\alpha(G) = \lfloor ({|V(G)|-k+1})/{2}\rfloor+\ell$; and $q$-tight for some integer $q\ge0$ if $\alpha(G) = \lfloor ({|V(G)|-k+1})/{2}\rfloor+\ell-q$. In this paper, we first prove that for all $k\geq 24$, the only tight $(k, 0)$-stable graphs are $K_{k+1}$ and $K_{k+2}$, answering a question of Dong and Luo [arXiv: 2401.16639]. We then prove that for all nonnegative integers $k, \ell, q$ with $k\geq 3\ell+3$, every $q$-tight $(k,\ell)$-stable graph has at most $k-3\ell-3+2^{3(\ell+2q+4)^2}$ vertices, answering a question of Dong and Luo in the negative., Comment: 11 pages

Interfacial ferroelectricity emerges in heterostructures consisting of nonpolar van der Waals (vdW) layers, greatly expanding the scope of two dimensional ferroelectrics. In particular, the unconventional moire ferroelectricity observed in bilayer graphene/boron nitride (BN) heterostructures, exhibits promising functionalities with topological current, superconductivity and synaptic responses. However, the debate about its mechanism - correlation driven charge transfer between two graphene layers - limits device reproducibility and hence large-scale production. Here by designing a single-layer graphene encapsulated by lattice-mismatched WSe2, we identify the ferroelectricity as stemming from - instead of graphene moire bands - the particular BN, where interfacial sliding ferroelectricity must play a role. With similar structures, multilayer twisted MoS2 is found to reproduce the ferroelectricity. The key is a conductive moire ferroelectric, where the screened gate and the pinned domain wall together result in unchanged electronic states, i.e. anomalous screening. The intimate connection to interfacial sliding ferroelectricity thus provides advantages of diverse choices of constituent materials and robust polarization switching while preserving the unique anomalous screening, paving the way to reproducible and reliable memory-based devices in artificial intelligence.

In this paper, we show that all simple outerplanar graphs $G$ with minimum degree at least $2$ and positive Lin-Lu-Yau Ricci curvature on every edge have maximum degree at most $9$. Furthermore, if $G$ is maximally outerplanar, then $G$ has at most $10$ vertices. Both upper bounds are sharp., Comment: 10 pages

In this paper, we show that the oriented diameter of any $n$-vertex $2$-connected near triangulation is at most $\lceil{\frac{n}{2}}\rceil$ (except for seven small exceptions), and the upper bound is tight. This extends a result of Wang et.al. on the oriented diameter of maximal outerplanar graphs, and improves an upper bound of $n/2+O(\sqrt{n})$ on the oriented diameter of planar triangulations by Mondal, Parthiban and Rajasingh.

Edge/fog computing, as a distributed computing paradigm, satisfies the low-latency requirements of ever-increasing number of IoT applications and has become the mainstream computing paradigm behind IoT applications. However, because large number of IoT applications require execution on the edge/fog resources, the servers may be overloaded. Hence, it may disrupt the edge/fog servers and also negatively affect IoT applications' response time. Moreover, many IoT applications are composed of dependent components incurring extra constraints for their execution. Besides, edge/fog computing environments and IoT applications are inherently dynamic and stochastic. Thus, efficient and adaptive scheduling of IoT applications in heterogeneous edge/fog computing environments is of paramount importance. However, limited computational resources on edge/fog servers imposes an extra burden for applying optimal but computationally demanding techniques. To overcome these challenges, we propose a Deep Reinforcement Learning-based IoT application Scheduling algorithm, called DRLIS to adaptively and efficiently optimize the response time of heterogeneous IoT applications and balance the load of the edge/fog servers. We implemented DRLIS as a practical scheduler in the FogBus2 function-as-a-service framework for creating an edge-fog-cloud integrated serverless computing environment. Results obtained from extensive experiments show that DRLIS significantly reduces the execution cost of IoT applications by up to 55%, 37%, and 50% in terms of load balancing, response time, and weighted cost, respectively, compared with metaheuristic algorithms and other reinforcement learning techniques.

Fine-tuning visual models has been widely shown promising performance on many downstream visual tasks. With the surprising development of pre-trained visual foundation models, visual tuning jumped out of the standard modus operandi that fine-tunes the whole pre-trained model or just the fully connected layer. Instead, recent advances can achieve superior performance than full-tuning the whole pre-trained parameters by updating far fewer parameters, enabling edge devices and downstream applications to reuse the increasingly large foundation models deployed on the cloud. With the aim of helping researchers get the full picture and future directions of visual tuning, this survey characterizes a large and thoughtful selection of recent works, providing a systematic and comprehensive overview of existing work and models. Specifically, it provides a detailed background of visual tuning and categorizes recent visual tuning techniques into five groups: prompt tuning, adapter tuning, parameter tuning, and remapping tuning. Meanwhile, it offers some exciting research directions for prospective pre-training and various interactions in visual tuning., Comment: 37 pages. Accepted to ACM CSUR

An accurate and explainable automatic monitoring system is critical for the safety of high efficiency energy conversion plants that operate under extreme working condition. Nonetheless, currently available data-driven monitoring systems often fall short in meeting the requirements for either high-accuracy or interpretability, which hinders their application in practice. To overcome this limitation, a data-driven approach, AttentionMixer, is proposed under a generalized message passing framework, with the goal of establishing an accurate and interpretable radiation monitoring framework for energy conversion plants. To improve the model accuracy, the first technical contribution involves the development of spatial and temporal adaptive message passing blocks, which enable the capture of spatial and temporal correlations, respectively; the two blocks are cascaded through a mixing operator. To enhance the model interpretability, the second technical contribution involves the implementation of a sparse message passing regularizer, which eliminates spurious and noisy message passing routes. The effectiveness of the AttentionMixer approach is validated through extensive evaluations on a monitoring benchmark collected from the national radiation monitoring network for nuclear power plants, resulting in enhanced monitoring accuracy and interpretability in practice.

Parameter-efficient transfer learning (PETL) is an emerging research spot aimed at inexpensively adapting large-scale pre-trained models to downstream tasks. Recent advances have achieved great success in saving storage costs for various pre-trained models by updating a small number of parameters instead of full tuning. However, we notice that most existing PETL methods still incur non-negligible latency during inference. In this paper, we propose a parameter-efficient and computational friendly adapter for giant vision models, called RepAdapter. Specifically, we first prove that common adaptation modules can also be seamlessly integrated into most giant vision models via our structural re-parameterization, thereby achieving zero-cost during inference. We then investigate the sparse design and effective placement of adapter structure, helping our RepAdaper obtain other advantages in terms of parameter efficiency and performance. To validate RepAdapter, we conduct extensive experiments on 27 benchmark datasets of three vision tasks, i.e., image and video classifications and semantic segmentation. Experimental results show the superior performance and efficiency of RepAdapter than the state-of-the-art PETL methods. For instance, RepAdapter outperforms full tuning by +7.2% on average and saves up to 25% training time, 20% GPU memory, and 94.6% storage cost of ViT-B/16 on VTAB-1k. The generalization ability of RepAdapter is also well validated by a bunch of vision models. Our source code is released at https://github.com/luogen1996/RepAdapter.

The spread of a graph $G$ is the difference between the largest and smallest eigenvalues of the adjacency matrix of $G$. In this paper, we consider the family of graphs which contain no $K_{2,t}$-minor. We show that for any $t\geq 2$, there is an integer $\xi_t$ such that the maximum spread of an $n$-vertex $K_{2,t}$-minor-free graph is achieved by the graph obtained by joining a vertex to the disjoint union of $\lfloor \frac{2n+\xi_t}{3t}\rfloor$ copies of $K_t$ and $n-1 - t\lfloor \frac{2n+\xi_t}{3t}\rfloor$ isolated vertices. The extremal graph is unique, except when $t\equiv 4 \mod 12$ and $\frac{2n+ \xi_t} {3t}$ is an integer, in which case the other extremal graph is the graph obtained by joining a vertex to the disjoint union of $\lfloor \frac{2n+\xi_t}{3t}\rfloor-1$ copies of $K_t$ and $n-1-t(\lfloor \frac{2n+\xi_t}{3t}\rfloor-1)$ isolated vertices. Furthermore, we give an explicit formula for $\xi_t$., Comment: Minor revisions. arXiv admin note: text overlap with arXiv:2209.13776

Enterprise relation extraction aims to detect pairs of enterprise entities and identify the business relations between them from unstructured or semi-structured text data, and it is crucial for several real-world applications such as risk analysis, rating research and supply chain security. However, previous work mainly focuses on getting attribute information about enterprises like personnel and corporate business, and pays little attention to enterprise relation extraction. To encourage further progress in the research, we introduce the CEntRE, a new dataset constructed from publicly available business news data with careful human annotation and intelligent data processing. Extensive experiments on CEntRE with six excellent models demonstrate the challenges of our proposed dataset.

The spread of a graph $G$ is the difference between the largest and smallest eigenvalue of the adjacency matrix of $G$. Gotshall, O'Brien and Tait conjectured that for sufficiently large $n$, the $n$-vertex outerplanar graph with maximum spread is the graph obtained by joining a vertex to a path on $n-1$ vertices. In this paper, we disprove this conjecture by showing that the extremal graph is the graph obtained by joining a vertex to a path on $\lceil (2n-1)/3\rceil$ vertices and $\lfloor(n-2)/3\rfloor$ isolated vertices. For planar graphs, we show that the extremal $n$-vertex planar graph attaining the maximum spread is the graph obtained by joining two nonadjacent vertices to a path on $\lceil(2n-2)/3\rceil$ vertices and $\lfloor(n-4)/3\rfloor$ isolated vertices.

Resource management is the principal factor to fully utilize the potential of Edge/Fog computing to execute real-time and critical IoT applications. Although some resource management frameworks exist, the majority are not designed based on distributed containerized components. Hence, they are not suitable for highly distributed and heterogeneous computing environments. Containerized resource management frameworks such as FogBus2 enable efficient distribution of framework's components alongside IoT applications' components. However, the management, deployment, health-check, and scalability of a large number of containers are challenging issues. To orchestrate a multitude of containers, several orchestration tools are developed. But, many of these orchestration tools are heavy-weight and have a high overhead, especially for resource-limited Edge/Fog nodes. Thus, for hybrid computing environments, consisting of heterogeneous Edge/Fog and/or Cloud nodes, lightweight container orchestration tools are required to support both resource-limited resources at the Edge/Fog and resource-rich resources at the Cloud. Thus, in this paper, we propose a feasible approach to build a hybrid and lightweight cluster based on K3s, for the FogBus2 framework that offers containerized resource management framework. This work addresses the challenge of creating lightweight computing clusters in hybrid computing environments. It also proposes three design patterns for the deployment of the FogBus2 framework in hybrid environments, including 1) Host Network, 2) Proxy Server, and 3) Environment Variable. The performance evaluation shows that the proposed approach improves the response time of real-time IoT applications up to 29% with acceptable and low overhead., Comment: 20 pages, 10 figures

Currently, due to the advantages of light weight, simple deployment, multi-environment support, short startup time, scalability, and easy migration, container technology has been widely used in both cloud and edge/fog computing, and addresses the problem of device heterogeneity in different computing environments. On this basis, as one of the most popular container orchestration and management systems, Kubernetes almost dominates the cloud environment. However, since it is primarily designed for centralized resource management scenarios where computing resources are sufficient, the system is unstable in edge environments due to hardware limitations. Therefore, in order to realize container orchestration in the cloud and edge/fog hybrid computing environment, we propose a feasible approach to build a hybrid clustering based on K3s, which solves the problem that virtual instances in different environments cannot be connected due to IP addresses. We also propose three design patterns for deploying the FogBus2 framework into hybrid environments, including 1) Host Network Mode, 2) Proxy Server, and 3) Environment Variable.

A graph $G$ is said to satisfy the Vizing bound if $\chi(G)\leq \omega(G)+1$, where $\chi(G)$ and $\omega(G)$ denote the chromatic number and clique number of $G$, respectively. It was conjectured by Randerath in 1998 that if $G$ is a triangle-free and fork-free graph, where the fork (also known as trident) is obtained from $K_{1,4}$ by subdividing two edges, then $G$ satisfies the Vizing bound. In this paper, we confirm this conjecture., Comment: 24 pages, 1 figure