Cross-domain sequential recommendation aims to generate accurate recommendations by leveraging users' historical interactions across domains. However, existing methods have two limitations: 1) When transferring user's preferences from the source domain, they encode preferences into a static and holistic representation, ignoring the rich information inherent in the dynamic evolution of user preferences over time; 2) They adopt a distribution-agnostic full-transfer strategy, failing to effectively limit the transfer degree of source-domain preferences according to different data distributions, which poses a risk of negative transfer. To address these issues, we propose the Dynamic Preference Transfer (DPT) model. Unlike existing methods, DPT places greater emphasis on the dynamic transfer of real-time preferences. First, DPT captures the causal features through the causal self-attention mechanism, and then realizes dynamic preference transfer at each time step via the causal cross-attention mechanism, thereby tracking the temporal dynamics of preferences from source domains. Second, to mitigate the negative transfer issue, a temperature-controlled mechanism is designed to adaptively balance source and target domain preferences, leveraging a temperature-controlled sigmoid function to effectively suppress interference from irrelevant preferences. Experimental results on multiple benchmark datasets show that the proposed method achieves significant performance improvements compared with the state-of-the-art (SOTA) methods, verifying its effectiveness and superiority. The codes are available in https://github.com/iryand/DPT.

Distant supervised relation extraction (DSRE) obtains large amounts of data cost-effectively by aligning knowledge base with natural texts but also brings noisy data. Existing methods deal with noise through multi-instance learning (MIL) with attention. However, these approaches typically use attention at sentence-level and above, while ignoring word-level information. Intuitively, words in the same sentence are also of different importance. However, effective methods for distinguishing word-level importance differences are still lacking. Because multiplying with weighted attention leads word embeddings to be shifted in vector space. Therefore, we proposed WMRE. Specifically, WMRE concatenates the embeddings of multiple sentences. Then it uses filtering attention to remove low-importance words at different levels to extract multi-hierarchical features. In WMRE, words are the smallest units of mining information, and filtering attention distinguishes word-level importance differences while avoiding the embedding shifts. Extensive experiments on two widely used datasets show that WMRE fully utilizes word-level information and achieves state-of-the-art (SOTA) on both datasets.

Since ultrasound imaging technique is convenient and real-time, it plays a crucial role in diagnosing thyroid nodules. With the development of deep learning, computer-aided diagnosis models have been widely applied to diagnose thyroid nodules, in which thyroid nodule segmentation is a basic yet essential task. Semi-supervised learning is a popular topic for thyroid ultrasound images segmentation under limited annotations. However, current mainstream semi-supervised methods (e.g., for nature image scenes) commonly produce poor segmentation results when facing thyroid ultrasound images. We analyze that there are two main reasons: First, since these methods lack the targeted learning for the ambiguous regions of images that may contain complementary clues for segmentation, the model will likely be over-fitting in the regions that are easy to predict, leading to fail to make full use of the unlabeled data. Second, these methods lack the shape constraint on thyroid nodules, resulting in incomplete segmentation shape on the nodule boundary. To address the above issues, we propose a novel Uncertainty-guided Dual Task Framework (UDTF). Concretely, we propose an Uncertainty Region Selection Module (URSM) to guide the segmentation task to learn from the ambiguous regions calculated by prototype under the constraint of consistency regularization. Additionally, to further keep the integrity of the nodules, we propose a Dual Task Module (DTM) to impose the shape constraint on thyroid nodules by exploring the task-level consistency between the segmentation and auxiliary reconstruction task. Extensive experiments are conducted on two thyroid ultrasound image datasets, including a private Philips Thyroid Ultrasound dataset and a public TN3K dataset. The results show that UDTF achieves superior performance compared to several other state-of-the-art methods.

Federated Learning has emerged as a promising machine learning paradigm to protect data privacy. However, the differences between heterogeneous clients and the performance bottleneck of central server limit the efficiency of FL. As a typical decentralized solution, the combination of blockchain and FL has been studied in recent years. However, the use of single-chain blockchain and traditional consensus algorithms in these studies have drawbacks such as high resource consumption, low TPS and low scalability. This paper proposes an efficient solution that combines a DAG blockchain and FL, called GCFL(Graph with Coordinator Federated Learning). GCFL introduces a new block structure that reduces data redundancy. For DAG blockchains, we proposed a two-phase tips selection consensus algorithm that can reduce resource consumption and tolerate a certain proportion of malicious nodes. Simulation experiments show that GCFL has higher stability and fast convergence time for targeted accuracy compared to traditional on-device FL systems.

With the maturity of artificial intelligence, AI-aided diagnosis technology is gradually widely applied in clinical medicine. However, for the same pathological tissue, medical images produced by different types of instruments usually possess different data distributions. Because of the domain shift phenomenon, AI-aided diagnosis cannot accurately diagnose medical images in other domains, which is a waste of precious medical images. This paper proposes a Multi-Scale Self-Attention Unsupervised Domain Adaptive framework (MSDAN), which consists of three modules. First, the multi-scale framework constrains the source domain features and target domain features by optimizing adversarial losses with different level features. Second, the mix-up discriminator extracts latent spatial features by mixing up source domain and target domain features. Finally, MSDAN learns the geometric information of the pathological tissues in medical images through the self-attention module, thereby improving the transfer effect of the semantic information in medical images. Extensive experiments prove that the proposed approach can achieve superior performance on tasks with various degrees of domain shift and data complexity, especially for thyroid ultrasound images.

For delay-sensitive applications in mobile edge computing (MEC), task admission approach is of vital importance, and there has been a lot of researches in this field. But previous works focus on the situation with only one MEC server that is a simplification of the real world. In multi-servers situation, some devices may be within the service range of multiple MEC servers, so that they could choose which MEC server to offload. We formulate this problem to a multiple-choice integer program (MCIP) and utilize Ben's genetic algorithm to solve it. The simulation results show that our approach can significantly reduce energy consumption and every task can catch its deadline under almost all experiments.

In this paper, we investigate a nonuniform node distribution strategy to mitigate the energy hole problem in wireless sensor networks (WSNs). Firstly, based on the analysis of energy consumption, we deduce a novel continuous node density function. Secondly, with the transformation of the density function, we obtain the disk radius function. And then, based on the disk radius function, we propose a novel nonuniform node distribution using adaptive Poisson disk (NDAPD). Our strategy can make the node density vary continuously in the network and mitigate the energy hole problem effectively. Finally, a routing algorithm tailored is presented for the proposed nonuniform node distribution. Compared with other well-known nonuniform node distribution strategies, NDAPD can reduce the number of network nodes effectively, improve the utilization of network energy by 3%-5% and increase the data transmission rate to 100% in most cases. Like other strategies, our strategy can be applied to most scenarios.

Most of the previous studies on the user preferences assume that there is a personal transitive preference ranking of the consumable media like images. For example, the transitivity of preferences is one of the most important assumptions in the recommender system research. However, the intransitive relations have also been widely observed, such as the win/loss relations in online video games, in sport matches, and even in rock-paper-scissors games. It is also found that different subjects demonstrate the personalized intransitive preferences in the pairwise comparisons between the applicants for college admission. Since the intransitivity of preferences on images has barely been studied before and has a large impact on the research of personalized image search and recommendation, it is necessary to propose a novel method to predict the personalized intransitive preferences of images. In this paper, we propose the novel Multi-Criterion preference (MuCri) models to predict the intransitive relations in the image preferences. The MuCri models utilize different kinds of image content features as well as the latent features of users and images. Meanwhile, a new data set is constructed in this paper, in order to evaluate the performance of the MuCri models. The experimental evaluation shows that the MuCri models outperform all the baselines. Due to the interdisciplinary nature of this topic, we believe it would widely attract the attention of researchers in the image processing community as well as in other communities, such as machine learning, multimedia, and recommender system.

Community detection has attracted great interest in graph analysis and mining during the past decade, and a great number of approaches have been developed to address this problem. However, the lack of a uniform framework and a reasonable evaluation method makes it a puzzle to analyze, compare and evaluate the extensive work, let alone picking out a best one when necessary. In this paper, we design a tool called CoDAR, which reveals the generalized procedure of community detection and monitors the real-time structural changes of network during the detection process. Moreover, CoDAR adopts 12 recognized metrics and builds a rating model for performance evaluation of communities to recom- mend the best-performing algorithm. Finally, the tool also provides nice interactive windows for display.

This paper presents an automatic method for computing an anisotropic 2D shape distribution on an arbitrary 2-manifold mesh. Our method allows the user to specify the direction as well as the density of the distribution. Using a pre-computed lookup table, our method can efficiently detect collision among the shapes to be distributed on the 3D mesh. In contrast to existing approaches, which usually assume the 2D objects are isotropic and have simple geometry, our method works for complex 2D objects and can guarantee the distribution is conflict-free, which is a critical constraint in many applications. It is able to compute multi-class shape distributions in parallel. Our method does not require global parameterization of the input 3D mesh. Instead, it computes local parameterizations on the fly using geodesic polar coordinates. Thanks to a recent breakthrough in geodesic computation, the local parameterization can be computed at low cost. As a result, our method can be applied to models with complicated geometry and topology. Experimental results on a wide range of 3D models and 2D anisotropic shapes demonstrate the good performance and effectiveness of our method.

Chinese calligraphy art is of significant importance in Chinese traditional culture, and meanwhile, the way to carry it forward in our information era is a critical issue. Thus, aiming at making a progress of the problem mentioned above, we present a novel method of writing Chinese calligraphy on arbitrary surfaces (triangle mesh). In this paper, Gong Qi calligraphy (i.e. Qi fonts) is chosen as test samples in the reason that it is one of the most famous calligraphy in China for its liquid structure and concise strokes. This paper consists of following four steps. Firstly, each character is decomposed to strokes, and we use the dynamic balance of bouncing disks approach to achieve the strokes' centerlines and corresponding radii which will be applied in vectorizing the strokes of characters through Disk B-Spline Curves. Secondly, both a fast geodesic algorithm and an exponential map method are employed which the former is to calculate the geodesic distance between every two vertexes of the character mapped region, while the latter is to obtain the geodesic coordinates corresponding to every vertexes of the triangle meshes in 3D, namely the geodesic triangulation. Thirdly, 3D points coordinates on surfaces, correspond to vectored character in tangent plane, are acquired on the basis of geodesic triangulation, thereby the character is able to be written on the surfaces. At last, some experiments are accomplished to test and verify the accuracy and efficiency of our method.

In many graphics applications, the computation of exact geodesic distance is very important. However, the high computational cost of existing geodesic algorithms means that they are not practical for large-scale models or time-critical applications. To tackle this challenge, we propose the Parallel Chen-Han (or PCH) algorithm, which extends the classic Chen-Han (CH) discrete geodesic algorithm to the parallel setting. The original CH algorithm and its variant both lack a parallel solution because the windows (a key data structure that carries the shortest distance in the wavefront propagation) are maintained in a strict order or a tightly coupled manner, which means that only one window is processed at a time. We propose dividing the CH's sequential algorithm into four phases, window selection, window propagation, data organization, and events processing so that there is no data dependence or conflicts in each phase and the operations within each phase can be carried out in parallel. The proposed PCH algorithm is able to propagate a large number of windows simultaneously and independently. We also adopt a simple yet effective strategy to control the total number of windows. We implement the PCH algorithm on modern GPUs (such as Nvidia GTX 580) and analyze the performance in detail. The performance improvement (compared to the sequential algorithms) is highly consistent with GPU double-precision performance (GFLOPS). Extensive experiments on real-world models demonstrate an order of magnitude improvement in execution time compared to the state-of-the-art.

This paper presents Texture Brush, an interactive interface for texturing 3D surfaces. We extend the conventional exponential map to a more general setting, in which the generator can be an arbitrary curve. Based on our extended exponential map, we develop a local parameterization method which naturally supports anisotropic texture mapping. With Texture Brush, the user can easily specify such local parameterization with a free-form stroke on the surface. We also propose a set of intuitive operations which are mainly based on 3D painting metaphor, including texture painting, texture cloning, texture animation design, and texture editing. Compared to the existing surface texturing techniques, our method enables a smoother and more natural work flow so that the user can focus on the design task itself without switching back and forth among different tools or stages. The encouraging experimental results and positive evaluation by artists demonstrate the efficacy of our Texture Brush for interactive texture mapping.

This paper presents a simple yet effective algorithm to improve an arbitrary Poisson disk sampling in R2 and R3 to reach the maximal property, i.e., no more Poisson disk can be inserted. Taking a non-maximal Poisson disk sampling as input, our algorithm efficiently detects the regions allowing additional samples and then generates Poisson disks in these regions. The key idea is to convert the complicated plane or space searching problem into a simple searching on circles or spheres, which is one dimensional lower than the original sampling domain. Our algorithm is memory efficient, fully parallel and highly fast by using modern graphics card.

This paper presents the Saddle Vertex Graph (SVG), a novel solution to the discrete geodesic problem. The SVG is a sparse undirected graph that encodes complete geodesic distance information: a geodesic path on the mesh is equivalent to a shortest path on the SVG, which can be solved efficiently using the shortest path algorithm (e.g., Dijkstra algorithm). The SVG method solves the discrete geodesic problem from a local perspective. We have observed that the polyhedral surface has some interesting and unique properties, such as the fact that the discrete geodesic exhibits a strong local structure, which is not available on the smooth surfaces. The richer the details and complicated geometry of the mesh, the stronger such local structure will be. Taking advantage of the local nature, the SVG algorithm breaks down the discrete geodesic problem into significantly smaller sub-problems, and elegantly enables information reuse. It does not require any numerical solver, and is numerically stable and insensitive to the mesh resolution and tessellation. Users can intuitively specify a model-independent parameter K , which effectively balances the SVG complexity and the accuracy of the computed geodesic distance. More importantly, the computed distance is guaranteed to be a metric. The experimental results on real-world models demonstrate significant improvement to the existing approximate geodesic methods in terms of both performance and accuracy.

Poisson disk sampling has excellent spatial and spectral properties, and plays an important role in a variety of visual computing. Although many promising algorithms have been proposed for multidimensional sampling in euclidean space, very few studies have been reported with regard to the problem of generating Poisson disks on surfaces due to the complicated nature of the surface. This paper presents an intrinsic algorithm for parallel Poisson disk sampling on arbitrary surfaces. In sharp contrast to the conventional parallel approaches, our method neither partitions the given surface into small patches nor uses any spatial data structure to maintain the voids in the sampling domain. Instead, our approach assigns each sample candidate a random and unique priority that is unbiased with regard to the distribution. Hence, multiple threads can process the candidates simultaneously and resolve conflicts by checking the given priority values. Our algorithm guarantees that the generated Poisson disks are uniformly and randomly distributed without bias. It is worth noting that our method is intrinsic and independent of the embedding space. This intrinsic feature allows us to generate Poisson disk patterns on arbitrary surfaces in IR(n). To our knowledge, this is the first intrinsic, parallel, and accurate algorithm for surface Poisson disk sampling. Furthermore, by manipulating the spatially varying density function, we can obtain adaptive sampling easily.

Computing discrete geodesics on polyhedral surfaces plays an important role in computer graphics. In contrast to the well-studied "single-source, all-destination" discrete geodesic problem, little progress has been reported to the all-pairs geodesic, i.e., computing the geodesic distance between arbitrary two points on the surface. To our knowledge, the existing all-pairs geodesic algorithms have very high computational cost, thus, can not be applied to real-world models, which usually contain thousands of vertices. In this paper, we propose an efficient algorithm to approximate the all-pairs geodesic on triangular meshes. The pre-processing step takes O(mn2 log n) time for the input mesh with n vertices and m samples, where m (≪ n) is specified by the user, usually between a few hundred and several thousand. In the query step, our algorithm can compute the approximate geodesic distance between arbitrary pair of points (not necessarily mesh vertices) in O(1) time. Furthermore, the geodesic path and the geodesic distance field can be approximated in linear time. Both theoretical analysis and experimental results on real-world models demonstrate that our algorithm is efficient and accurate. We demonstrate the efficacy of our algorithm on the interactive texture mapping by using discrete exponential map.

Geodesic plays an important role in geometric computation and analysis. Rather than the widely studied single source all destination discrete geodesic problem, very little work has been reported on the all pairs geodesic distance query So far, the best known result is due to Cook IV and Wenk [2009], who pre-computed the pairwise geodesic between any two mesh vertices in O(n52α(n) logn) time complexity and O(n4) space complexity, where n is the number of mesh vertices and α(n) the inverse Ackermann function. Then the geodesic distance between any pair of points on the mesh edges can be computed in O(m + logn) time, where m is the number of edges crossed by the geodesic path. Although Cook IV and Wenk's algorithm is able to compute the exact geodesic the high computational cost limits its applications to real-world models which usually contain thousands of vertices.