检索结果-内蒙古大学图书馆

Warm up cold-start advertisements: Improving CTR predictions via learning to learn ID embeddings

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Pan, Feiyang Li, Shuokai Ao, Xiang Tang, Pingzhong He, Qing Institute of Computing Technology Chinese Academy of Sciences IIIS Tsinghua University Key Lab of Intelligent Information Processing of Chinese Academy of Sciences University of Chinese Academy of Sciences China

Click-through rate (CTR) prediction has been one of the most central problems in computational advertising. Lately, embedding techniques that produce low-dimensional representations of ad IDs drastically improve CTR prediction accuracies. However, such learning techniques are data demanding and work poorly on new ads with little logging data, which is known as the cold-start problem. In this paper, we aim to improve CTR predictions during both the cold-start phase and the warm-up phase when a new ad is added to the candidate pool. We propose Meta-Embedding, a meta-learningbased approach that learns to generate desirable initial embeddings for new ad IDs. The proposed method trains an embedding generator for new ad IDs by making use of previously learned ads through gradient-based meta-learning. In other words, our method learns how to learn better embeddings. When a new ad comes, the trained generator initializes the embedding of its ID by feeding its contents and attributes. Next, the generated embedding can speed up the model fitting during the warm-up phase when a few labeled examples are available, compared to the existing initialization methods. Experimental results on three real-world datasets showed that Meta-Embedding can significantly improve both the cold-start and warm-up performances for six existing CTR prediction models, ranging from lightweight models such as Factorization Machines to complicated deep models such as PNN and DeepFM. All of the above apply to conversion rate (CVR) predictions as well. Copyright © 2019, The Authors. All rights reserved.

关键词： Embeddings

Global-Local Bidirectional Reasoning for Unsupervised Representation Learning of 3D Point Clouds

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Rao, Yongming Lu, Jiwen Zhou, Jie Department of Automation Tsinghua University China State Key Lab of Intelligent Technologies and Systems China Beijing National Research Center for Information Science and Technology China Tsinghua Shenzhen International Graduate School Tsinghua University China

Local and global patterns of an object are closely related. Although each part of an object is incomplete, the underlying attributes about the object are shared among all parts, which makes reasoning the whole object from a single part possible. We hypothesize that a powerful representation of a 3D object should model the attributes that are shared between parts and the whole object, and distinguishable from other objects. Based on this hypothesis, we propose to learn point cloud representation by bidirectional reasoning between the local structures at different abstraction hierarchies and the global shape without human supervision. Experimental results on various benchmark datasets demonstrate the unsupervisedly learned representation is even better than supervised representation in discriminative power, generalization ability, and robustness. We show that unsupervisedly trained point cloud models can outperform their supervised counterparts on downstream classification tasks. Most notably, by simply increasing the channel width of an SSG PointNet++1, our unsupervised model surpasses the state-of-the-art supervised methods on both synthetic and real-world 3D object classification datasets. We expect our observations to offer a new perspective on learning better representation from data structures instead of human annotations for point cloud understanding.2 Copyright © 2020, The Authors. All rights reserved.

关键词： Classification (of information)

Deep face super-resolution with iterative collaboration between attentive recovery and landmark estimation

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Ma, Cheng Jiang, Zhenyu Rao, Yongming Lu, Jiwen Zhou, Jie Department of Automation Tsinghua University China State Key Lab of Intelligent Technologies and Systems China Beijing National Research Center for Information Science and Technology China Tsinghua Shenzhen International Graduate School Tsinghua University China

Recent works based on deep learning and facial priors have succeeded in super-resolving severely degraded facial images. However, the prior knowledge is not fully exploited in existing methods, since facial priors such as landmark and component maps are always estimated by low-resolution or coarsely super-resolved images, which may be inaccurate and thus affect the recovery performance. In this paper, we propose a deep face super-resolution (FSR) method with iterative collaboration between two recurrent networks which focus on facial image recovery and landmark estimation respectively. In each recurrent step, the recovery branch utilizes the prior knowledge of landmarks to yield higher-quality images which facilitate more accurate landmark estimation in turn. Therefore, the iterative information interaction between two processes boosts the performance of each other progressively. Moreover, a new attentive fusion module is designed to strengthen the guidance of landmark maps, where facial components are generated individually and aggregated attentively for better restoration. Quantitative and qualitative experimental results show the proposed method significantly outperforms state-of-the-art FSR methods in recovering high-quality face images. Copyright © 2020, The Authors. All rights reserved.

关键词： Recovery

Deep Face Super-Resolution With Iterative Collaboration Between Attentive Recovery and Landmark Estimation

学校读者我要写书评

暂无评论

Deep Face Super-Resolution With Iterative Collaboration Betw...

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Cheng Ma Zhenyu Jiang Yongming Rao Jiwen Lu Jie Zhou Department of Automation Tsinghua University China State Key Lab of Intelligent Technologies and Systems China Beijing National Research Center for Information Science and Technology China Tsinghua Shenzhen International Graduate School Tsinghua University China

ISBN: (数字)9781728171685

ISBN: (纸本)9781728171692

关键词： Face Image resolution Feature extraction Collaboration Heating systems Estimation Iterative methods

Bidet: An efficient binarized object detector

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Wang, Ziwei Wu, Ziyi Lu, Jiwen Zhou, Jie Department of Automation Tsinghua University China State Key Lab of Intelligent Technologies and Systems China Beijing National Research Center for Information Science and Technology China Tsinghua Shenzhen International Graduate School Tsinghua University China

In this paper, we propose a binarized neural network learning method called BiDet for efficient object detection. Conventional network binarization methods directly quantize the weights and activations in one-stage or twostage detectors with constrained representational capacity, so that the information redundancy in the networks causes numerous false positives and degrades the performance significantly. On the contrary, our BiDet fully utilizes the representational capacity of the binary neural networks for object detection by redundancy removal, through which the detection precision is enhanced with alleviated false positives. Specifically, we generalize the information bottleneck (IB) principle to object detection, where the amount of information in the high-level feature maps is constrained and the mutual information between the feature maps and object detection is maximized. Meanwhile, we learn sparse object priors so that the posteriors are concentrated on informative detection prediction with false positive elimination. Extensive experiments on the PASCAL VOC and COCO datasets show that our method outperforms the state-of-theart binary neural networks by a sizable margin. Copyright © 2020, The Authors. All rights reserved.

关键词： Object recognition

PV-RAFT: Point-voxel correlation fields for scene flow estimation of point clouds

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Wei, Yi Wang, Ziyi Rao, Yongming Lu, Jiwen Zhou, Jie Department of Automation Tsinghua University China State Key Lab of Intelligent Technologies and Systems China Beijing National Research Center for Information Science and Technology China Tsinghua Shenzhen International Graduate School Tsinghua University China

In this paper, we propose a Point-Voxel Recurrent All-Pairs Field Transforms (PV-RAFT) method to estimate scene flow from point clouds. Since point clouds are irregular and unordered, it is challenging to efficiently extract features from all-pairs fields in the 3D space, where all-pairs correlations play important roles in scene flow estimation. To tackle this problem, we present point-voxel correlation fields, which capture both local and long-range dependencies of point pairs. To capture point-based correlations, we adopt the K-Nearest Neighbors search that preserves fine-grained information in the local region. By voxelizing point clouds in a multi-scale manner, we construct pyramid correlation voxels to model long-range correspondences. Integrating these two types of correlations, our PV-RAFT makes use of all-pairs relations to handle both small and large displacements. We evaluate the proposed method on the FlyingThings3D and KITTI Scene Flow 2015 datasets. Experimental results show that PV-RAFT outperforms state-of-the-art methods by remarkable margins. Copyright © 2020, The Authors. All rights reserved.

关键词： Nearest neighbor search

Global-Local Bidirectional Reasoning for Unsupervised Representation Learning of 3D Point Clouds

学校读者我要写书评

暂无评论

Global-Local Bidirectional Reasoning for Unsupervised Repres...

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Yongming Rao Jiwen Lu Jie Zhou Department of Automation Tsinghua University China State Key Lab of Intelligent Technologies and Systems China Beijing National Research Center for Information Science and Technology China Tsinghua Shenzhen International Graduate School Tsinghua University China

ISBN: (数字)9781728171685

ISBN: (纸本)9781728171692

Local and global patterns of an object are closely related. Although each part of an object is incomplete, the underlying attributes about the object are shared among all parts, which makes reasoning the whole object from a single part possible. We hypothesize that a powerful representation of a 3D object should model the attributes that are shared between parts and the whole object, and distinguishable from other objects. Based on this hypothesis, we propose to learn point cloud representation by bidirectional reasoning between the local structures at different abstraction hierarchies and the global shape without human supervision. Experimental results on various benchmark datasets demonstrate the unsupervisedly learned representation is even better than supervised representation in discriminative power, generalization ability, and robustness. We show that unsupervisedly trained point cloud models can outperform their supervised counterparts on downstream classification tasks. Most notably, by simply increasing the channel width of an SSG PointNet++, our unsupervised model surpasses the state-of-the-art supervised methods on both synthetic and real-world 3D object classification datasets. We expect our observations to offer a new perspective on learning better representation from data structures instead of human annotations for point cloud understanding.

关键词： Three-dimensional displays Cognition Semantics Task analysis Feature extraction Shape Measurement

Deep learning for design metamaterial electromagnetic induction transparent device

学校读者我要写书评

暂无评论

Journal of Physics: Conference Series 2021年第1期1775卷

作者： Ziming Wei Zeyulin Zhang Wei Huang Shan Yin Wentao Zhang Guangxi Key Lab. of Optoelectronic Information Processing School of Electronic Engineering and Automation Guilin University of Electronic Technology Guilin Guangxi 541004 China

In this paper, we propose a deep learning model that can be used to reverse design metamaterial electromagnetic induction transparent (EIT) devices. This is a problem that is difficult to achieve with traditional numerical calculation methods. We use the coordinates of six specific points on the EIT transmission spectrum as the input of the neural network, and then the network can predict the structural parameters of the corresponding EIT device. We cite an example to prove that our method can be used to efficiently reverse design the structure of EIT devices. We believe that this method will open up a new way for the structural design of EIT devices and has great potential for expanding the application of terahertz EIT metamaterials.

关键词：

Optical fiber based soft curvature sensor with polyvinyl chloride reinforced silicone rubber substrate

学校读者我要写书评

暂无评论

Optik 2019年 178卷 567-574页

作者： Xiongxiong Zhang Yanming Song Fanyong Meng Guangkai Sun Lianqing Zhu Beijing Key Laboratory of Optoelectronic Measurement Technology Beijing Information Science & Technology University Beijing 100101 China Bionic and Intelligent Equipment Lab Beijing Information Science and Technology University Beijing 100016 China

To meet the demands for curvature measurement in the field of biomedical intelligent equipment, soft robots and complex structures of major infrastructures, an optical fiber based soft curvature sensor is proposed. The major contribution is obtaining a soft curvature sensor with extraordinary high sensitivity and repeatability. A silicone rubber sheet is glued on the surface of polyvinyl chloride to form polyvinyl chloride (PVC) reinforced silicone rubber substrate, the fiber Bragg grating is embedded into the substrate to form the sensor. The sensor is calibrated using standard curvature calibration blocks. The characteristic of the reflected spectrum of the sensor is analyzed. The relation between the sensitivity and the fiber embedded depth is obtained. The research results prove that the real-time measurement of structure curvature can be realized by the sensor. The polyvinyl chloride reinforced silicone rubber substrate is helpful to improve the sensitivity and repeatability of the sensor. The maximum sensitivity is up to 329.2pm/m −1 . The sensor has good consistency in repeated measurements, it has application presents in measuring the curvature of flexible mechanisms and complex structures.

关键词： Curvature sensor Composite substrate FBG sensing Measuring calibration