检索结果-内蒙古大学图书馆

CLIP-Driven Open-Vocabulary 3D Scene Graph Generation via Cross-Modality Contrastive Learning

学校读者我要写书评

暂无评论

CLIP-Driven Open-Vocabulary 3D Scene Graph Generation via Cr...

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Lianggangxu Chen Xuejiao Wang Jiale Lu Shaohui Lin Changbo Wang Gaoqi He School of Computer Science and Technology East China Normal University Shanghai China Key Laboratory of Advanced Theory and Application in Statistics and Data Science Ministry of Education China Chongqing Key Laboratory of Precision Optics Chongqing Institute of East China Normal University Chongqing China

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

3D Scene Graph Generation (3DSGG) aims to classify objects and their predicates within 3D point cloud scenes. However, current 3DSGG methods struggle with two main challenges. 1) The dependency on labor-intensive ground-truth annotations. 2) Closed-set classes training hampers the recognition of novel objects and predicates. Addressing these issues, our idea is to extract cross-modality features by CLIP from text and image data naturally related to 3D point clouds. Cross-modality features are used to train a robust 3D scene graph (3DSG)feature extractor. Specifically, we propose a novel Cross-Modality Contrastive Learning 3DSGG (CCL-3DSGG) method. Firstly, to align the text with 3DSG, the text is parsed into word level that are consistent with the 3DSG annotation. To enhance robustness during the alignment, adjectives are exchanged for different objects as negative samples. Then, to align the image with 3DSG, the camera view is treated as a positive sample and other views as negatives. Lastly, the recognition of novel object and predicate classes is achieved by calculating the cosine similarity between prompts and 3DSG features. Our rigorous experiments confirm the superior open-vocabulary capability and applicability of CCL-3DSGG in real-world contexts.

关键词： Point cloud compression Training Computer vision Three-dimensional displays Annotations Contrastive learning Feature extraction

Assessment of Effects of Age and Gender on the Incubation Period of COVID-19 with a Mixture Regression Model

学校读者我要写书评

暂无评论

Journal of data science 2021年第1期19卷 p1-16页

作者： SIMING ZHENG JING QIN YONG ZHOU Academy of Mathematics and Systems Science University of Chinese Academy of Sciences Beijing China National Institute of Allergy and Infectious Diseases National Institutes of Health Bethesda Maryland USA Key Laboratory of Advanced Theory and Application in Statistics and Data Science MOE Academy of Statistics and Interdisciplinary Sciences Faculty of Economics and Management East China Normal University Shanghai China

Following the outbreak of COVID-19, various containment measures have been taken, including the use of quarantine. At present, the quarantine period is the same for everyone, since it is implicitly assumed that the incubation period distribution of COVID-19 is the same regardless of age or gender. For testing the effects of age and gender on the incubation period of COVID-19, a novel two-component mixture regression model is proposed. An expectation-maximization (EM) algorithm is adopted to obtain estimates of the parameters of interest, and the simulation results show that the proposed method outperforms the simple regression method and has robustness. The proposed method is applied to a Zhejiang COVID-19 dataset, and it is found that age and gender statistically have no effect on the incubation period of COVID-19, which indicates that the quarantine measure currently in operation is reasonable. [ABSTRACT FROM AUTHOR]

关键词： EM algorithm incubation period length-biased data mixture model

A General and Efficient Training for Transformer via Token Expansion

学校读者我要写书评

暂无评论

A General and Efficient Training for Transformer via Token E...

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Wenxuan Huang Yunhang Shen Jiao Xie Baochang Zhang Gaoqi He Ke Li Xing Sun Shaohui Lin School of Computer Science and Technology East China Normal University Shanghai China Tencent Youtu Lab China Xiamen University China Beihang University China Key Laboratory of Advanced Theory and Application in Statistics and Data Science - MOE China

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

The remarkable performance of Vision Transformers (ViTs) typically requires an extremely large training cost. Existing methods have attempted to accelerate the training of ViTs, yet typically disregard method universality with accuracy dropping. Meanwhile, they break the training consistency of the original transformers, including the consistency of hyperparameters, architecture, and strategy, which prevents them from being widely applied to different Transformer networks. In this paper, we propose a novel token growth scheme Token Expansion (termed ToE) to achieve consistent training acceleration for ViTs. We introduce an “initialization-expansion-merging” pipeline to maintain the integrity of the intermediate feature distribution of original transformers, preventing the loss of crucial learnable information in the training process. ToE can not only be seamlessly integrated into the training and fine-tuning process of transformers (e.g., DeiT and LV-ViT), but also effective for efficient training frameworks (e.g., EfficientTrain), without twisting the original training hyperparameters, architecture, and introducing additional training strategies. Extensive experiments demonstrate that ToE achieves about 1.3× faster for the training of ViTs in a lossless manner, or even with performance gains over the full-token training baselines. Code is available at https://***/Osilly/TokenExpansion.

关键词： Training Computer vision Accuracy Costs Codes Pipelines Computer architecture

A General and Efficient Training for Transformer via Token Expansion

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Huang, Wenxuan Shen, Yunhang Xie, Jiao Zhang, Baochang He, Gaoqi Li, Ke Sun, Xing Lin, Shaohui School of Computer Science and Technology East China Normal University Shanghai China Tencent Youtu Lab China Xiamen University China Beihang University China Key Laboratory of Advanced Theory and Application in Statistics and Data Science - MOE China

The remarkable performance of Vision Transformers (ViTs) typically requires an extremely large training cost. Existing methods have attempted to accelerate the training of ViTs, yet typically disregard method universality with accuracy dropping. Meanwhile, they break the training consistency of the original transformers, including the consistency of hyper-parameters, architecture, and strategy, which prevents them from being widely applied to different Transformer networks. In this paper, we propose a novel token growth scheme Token Expansion (termed ToE) to achieve consistent training acceleration for ViTs. We introduce an "initialization-expansion-merging" pipeline to maintain the integrity of the intermediate feature distribution of original transformers, preventing the loss of crucial learnable information in the training process. ToE can not only be seamlessly integrated into the training and fine-tuning process of transformers (e.g., DeiT and LV-ViT), but also effective for efficient training frameworks (e.g., EfficientTrain), without twisting the original training hyper-parameters, architecture, and introducing additional training strategies. Extensive experiments demonstrate that ToE achieves about 1.3× faster for the training of ViTs in a lossless manner, or even with performance gains over the full-token training baselines. Code is available at https://***/Osilly/TokenExpansion. Copyright © 2024, The Authors. All rights reserved.

关键词： Machine learning

A majorized-generalized alternating direction method of multipliers for convex composite programming

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Qin, Congying Xiao, Yunhai Li, Peili School of Mathematics and Statistics Henan University Kaifeng475000 China Henan Engineering Research Center for Artificial Intelligence Theory and Algorithms Henan University Kaifeng475000 China School of Statistics Key Laboratory of Advanced Theory and Application in Statistics and Data Science-MOE East China Normal University Shanghai200062 China

The linearly constrained convex composite programming problems whose objective function contains two blocks with each block being the form of nonsmooth+smooth arises frequently in multiple fields of applications. If both of the smooth terms are quadratic, this problem can be solved efficiently by using the symmetric Gaussian-Seidel (sGS) technique based proximal alternating direction method of multipliers (ADMM). However, in the non-quadratic case, the sGS technique can not be used any more, which leads to the separable structure of nonsmooth+smooth had to be ignored. In this paper, we present a generalized ADMM and particularly use a majorization technique to make the corresponding subproblems more amenable to efficient computations. Under some appropriate conditions, we prove its global convergence for the relaxation factor in (0, 2). We apply the algorithm to solve a kind of simulated convex composite optimization problems and a type of sparse inverse covariance matrix estimation problems which illustrates that the effectiveness of the algorithm are obvious. Copyright © 2021, The Authors. All rights reserved.

关键词： Covariance matrix

Cross Attention Based Style Distribution for Controllable Person Image Synthesis

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhou, Xinyue Yin, Mingyu Chen, Xinyuan Sun, Li Gao, Changxin Li, Qingli Shanghai Key Laboratory of Multidimensional Information Processing East China Normal University Shanghai China Key Laboratory of Advanced Theory and Application in Statistics and Data Science East China Normal University Shanghai China Shanghai AI Laboratory Shanghai China Huazhong University of Science and Technology Wuhan China

Controllable person image synthesis task enables a wide range of applications through explicit control over body pose and appearance. In this paper, we propose a cross attention based style distribution module that computes between the source semantic styles and target pose for pose transfer. The module intentionally selects the style represented by each semantic and distributes them according to the target pose. The attention matrix in cross attention expresses the dynamic similarities between the target pose and the source styles for all semantics. Therefore, it can be utilized to route the color and texture from the source image, and is further constrained by the target parsing map to achieve a clearer objective. At the same time, to encode the source appearance accurately, the self attention among different semantic styles is also added. The effectiveness of our model is validated quantitatively and qualitatively on pose transfer and virtual try-on tasks. Codes are available at https://***/xyzhouo/CASD. Copyright © 2022, The Authors. All rights reserved.

关键词： Semantics

NOVEL ENSEMBLE ALGORITHMS FOR RANDOM TWO-DOMAIN PARABOLIC PROBLEMS

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Sun, Yizhong Wang, Jiangshan Zheng, Haibiao School of Mathematical Sciences East China Normal University Shanghai China School of Mathematical Sciences East China Normal University Shanghai Key Laboratory of Pure Mathematics and Mathematical Practice Key Laboratory of Advanced Theory and Application in Statistics and Data Science Shanghai China

In this paper, three efficient ensemble algorithms are proposed for fast-solving the random fluid-fluid interaction model. Such a model can be simplified as coupling two heat equations with random diffusion coefficients and a friction parameter due to its complexity and uncertainty. We utilize the Monte Carlo method for the coupled model with random inputs to derive some deterministic fluid-fluid numerical models and use the ensemble idea to realize the fast computation of multiple problems. Our remarkable feature of these algorithms is employing the same coefficient matrix for multiple linear systems, significantly reducing the computational cost. By data-passing partitioned techniques, we can decouple the numerical models into two smaller sub-domain problems and achieve parallel computation. Theoretically, we derive that both algorithms are unconditionally stable and convergent. Finally, numerical experiments are conducted not only to support the theoretical results but also to validate the exclusive feature of the proposed algorithms. © 2023, CC BY-NC-SA.

关键词： Numerical models

Time-Consistent Portfolio Selection for Rank-Dependent Utilities in an Incomplete Market

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wei, Jiaqin Xia, Jianming Zhao, Qian Key Laboratory of Advanced Theory and Application in Statistics and Data Science-MOE School of Statistics East China Normal University Shanghai200062 China Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing100190 China School of Statistics and Information Shanghai University of International Business and Economics Shanghai201620 China

We investigate the portfolio selection problem for an agent with rank-dependent utility in an incomplete financial market. For a constant-coefficient market and CRRA utilities, we characterize the deterministic strict equilibrium strategies. In the case of time-invariant probability weighting function, we provide a comprehensive characterization of the deterministic strict equilibrium strategy. The unique non-zero equilibrium, if exists, can be determined by solving an autonomous ODE. In the case of time-variant probability weighting functions, we observe that there may be infinitely many non-zero deterministic strict equilibrium strategies, which are derived from the positive solutions to a nonlinear singular ODE. By specifying the maximal solution to the singular ODE, we are able to identify all the positive solutions. In addition, we address the issue of selecting an optimal strategy from the numerous equilibrium strategies available. © 2024, CC BY-NC-ND.

关键词： Commerce

An asymptotic analysis of minibatch-based momentum methods for linear regression models

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Gao, Yuan Zhu, Xuening Qi, Haobo Li, Guodong Zhang, Riquan Wang, Hansheng School of Statistics East China Normal University Shanghai China Key Laboratory of Advanced Theory and Application in Statistics and Data Science - MOE East China Normal University Shanghai China School of Data Science Fudan University Shanghai China Guanghua School of Management Peking University Beijing China Department of Statistics and Actuarial Science University of Hong Kong Hong Kong

Momentum methods have been shown to accelerate the convergence of the standard gradient descent algorithm in practice and theory. In particular, the minibatch-based gradient descent methods with momentum (MGDM) are widely used to solve large-scale optimization problems with massive datasets. Despite the success of the MGDM methods in practice, their theoretical properties are still underexplored. To this end, we investigate the theoretical properties of MGDM methods based on the linear regression models. We first study the numerical convergence properties of the MGDM algorithm and further provide the theoretically optimal tuning parameters specification to achieve faster convergence rate. In addition, we explore the relationship between the statistical properties of the resulting MGDM estimator and the tuning parameters. Based on these theoretical findings, we give the conditions for the resulting estimator to achieve the optimal statistical efficiency. Finally, extensive numerical experiments are conducted to verify our theoretical results. © 2021, CC BY.

关键词： Gradient methods