Filters








178,631 Hits in 4.9 sec

MLAN: Multi-Level Adversarial Network for Domain Adaptive Semantic Segmentation [article]

Jiaxing Huang, Dayan Guan, Shijian Lu, Aoran Xiao
2022 arXiv   pre-print
However, most adversarial learning based methods align source and target distributions at a global image level but neglect the inconsistency around local image regions.  ...  This paper presents a novel multi-level adversarial network (MLAN) that aims to address inter-domain inconsistency at both global image level and local region level optimally.  ...  ) and Nanyang Technological University (NTU) that is funded by the Singapore Government through the Industry Alignment Fund -Industry Collaboration Projects Grant. 14  ... 
arXiv:2103.12991v2 fatcat:ordpzyl5gnepfhyifjgbrhfk3q

Computation Offloading in Beyond 5G Networks: A Distributed Learning Framework and Applications [article]

Xianfu Chen and Celimuge Wu and Zhi Liu and Ning Zhang and Yusheng Ji
2020 arXiv   pre-print
Facing the trend of merging wireless communications and multi-access edge computing (MEC), this article studies computation offloading in the beyond fifth-generation networks.  ...  To address the technical challenges originating from the uncertainties and the sharing of limited resource in an MEC system, we formulate the computation offloading problem as a multi-agent Markov decision  ...  Across the time, each agent shares the local state information with all Feedbacks Surrounding Environment Joint Action Next Global State Current Global State Local States Local State Local  ... 
arXiv:2007.08001v1 fatcat:pomlqitbbfcxlp2byq2uowbhja

BEyond observation: an approach for ObjectNav [article]

Daniel V. Ruiz, Eduardo Todt
2021 arXiv   pre-print
It composes a multi-disciplinary field of robotics that encompasses embedded systems, control theory, path planning, Simultaneous Localization and Mapping (SLAM), scene reconstruction, and pattern recognition  ...  In this work, we present our exploratory research of how sensor data fusion and state-of-the-art machine learning algorithms can perform the Embodied Artificial Intelligence (E-AI) task called Visual Semantic  ...  and imitation learning.  ... 
arXiv:2106.11379v1 fatcat:abvvyxc53rgk5mvtx2qwkhohde

Horizontal and Vertical Reinforcement in Global Climate Governance

Martin Jänicke
2015 Energies  
in multi-level systems of governance.  ...  This paper is dealing with mechanisms that can accelerate the global diffusion of climate-friendly technologies.  ...  Multi-level governance-from the global to the local level-can be regarded as a general mechanism of reinforcement.  ... 
doi:10.3390/en8065782 fatcat:3rse7hknpzclvo5quqortnekpi

Unsupervised Multi-Modal Medical Image Registration via Discriminator-Free Image-to-Image Translation [article]

Zekang Chen, Jia Wei, Rui Li
2022 arXiv   pre-print
target modality.  ...  In clinical practice, well-aligned multi-modal images, such as Magnetic Resonance (MR) and Computed Tomography (CT), together can provide complementary information for image-guided therapies.  ...  Acknowledgments This work is supported in part by the Natural Science Foundation of Guangdong Province (2020A1515010717), NSF-1850492 and NSF-2045804.  ... 
arXiv:2204.13656v1 fatcat:ukegvxqypzae5l4nepagexjn2u

Beyond Greedy Search: Tracking by Multi-Agent Reinforcement Learning-based Beam Search [article]

Xiao Wang, Zhe Chen, Bo Jiang, Jin Tang, Bin Luo, Dacheng Tao
2022 arXiv   pre-print
Accordingly, this paper introduces a novel multi-agent reinforcement learning based beam search tracking strategy, termed BeamTracking.  ...  To track the target in a video, current visual trackers usually adopt greedy search for target object localization in each frame, that is, the candidate region with the maximum response score will be selected  ...  Index Terms-Visual Tracking, Multi-Agent Reinforcement Learning, Beam Search, Local and Global Search, Greedy Search I.  ... 
arXiv:2205.09676v2 fatcat:pcblgx4amja63ioid4ebx6bs5u

RePre: Improving Self-Supervised Vision Transformer with Reconstructive Pre-training [article]

Luya Wang, Feng Liang, Yangguang Li, Honggang Zhang, Wanli Ouyang, Jing Shao
2022 arXiv   pre-print
However, the dominant method, contrastive learning, mainly relies on an instance discrimination pretext task, which learns a global understanding of the image.  ...  This paper incorporates local feature learning into self-supervised vision transformers via Reconstructive Pre-training (RePre).  ...  In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2016. [Devlin et al., 2018]  ... 
arXiv:2201.06857v2 fatcat:alqu2db7xngmphbc23w7gwoobi

SGoLAM: Simultaneous Goal Localization and Mapping for Multi-Object Goal Navigation [article]

Junho Kim, Eun Sun Lee, Mingi Lee, Donsu Zhang, Young Min Kim
2021 arXiv   pre-print
We present SGoLAM, short for simultaneous goal localization and mapping, which is a simple and efficient algorithm for Multi-Object Goal navigation.  ...  Nonetheless, our approach performs on par with the state-of-the-art learning-based approaches. SGoLAM is ranked 2nd in the CVPR 2021 MultiON (Multi-Object Goal Navigation) challenge.  ...  The global map is further used in the Policy module. Goal Localization Given an RGB image I and depth map D, the Goal Localization module caches the location of target objects.  ... 
arXiv:2110.07171v1 fatcat:i5goy2gqqvdl5b7kws3ohvraie

Transfer of Knowledge through Collective Learning

Mohammad Rostami
2017 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Then I focus on a multi-agent learning setting, where related tasks are learned collectively and/or address privacy concerns.  ...  In my thesis, I explore this problem for knowledge transfer for multi-agent multi-task learning in a life-long learning paradigm.  ...  Beyond providing the substantial benefit of collective learning to our distributed learning problem, this effort will also address fundamental problems in distributed multi-task and transfer learning of  ... 
doi:10.1609/aaai.v31i1.10528 fatcat:coel2ajijnaavlcbexkjybetpm

Multi-level Domain Adaptive learning for Cross-Domain Detection [article]

Rongchang Xie, Fei Yu, Jiachao Wang, Yizhou Wang, Li Zhang
2019 arXiv   pre-print
To solve this problem, we propose a multi-level domain adaptive model to simultaneously align the distributions of local-level features and global-level features.  ...  Previous works use adversarial training to align global features across the domain shift and to achieve image information transfer.  ...  The global features are aligned well in both DA model and the proposed Multi-level DA model. However, the local features between source and target domain are mismatched in DA model.  ... 
arXiv:1907.11484v2 fatcat:sdj37szt35bpzcj3m4ohsu2cxy

Multi-Level Domain Adaptive Learning for Cross-Domain Detection

Rongchang Xie, Fei Yu, Jiachao Wang, Yizhou Wang, Li Zhang
2019 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)  
To solve this problem, we propose a multi-level domain adaptive model to simultaneously align the distributions of local-level features and global-level features.  ...  Previous works use adversarial training to align global features across the domain shift and to achieve image information transfer.  ...  The global features are aligned well in both DA model and the proposed Multi-level DA model. However, the local features between source and target domain are mismatched in DA model.  ... 
doi:10.1109/iccvw.2019.00401 dblp:conf/iccvw/XieYWWZ19 fatcat:ahytgtpfyfe4lbql25nzjrgem4

Local Patch Network with Global Attention for Infrared Small Target Detection [article]

Fang Chen, Chenqiang Gao, Fangcen Liu, Yue Zhao, Yuxi Zhou, Deyu Meng, Wangmeng Zuo
2021 arXiv   pre-print
By leveraging both the global and local properties, the data-driven framework proposed in this paper has fused multi-scale features for small target detection.  ...  A local patch network (LPNet) with global attention is proposed in this paper to detect small targets by jointly considering the global and local properties of infrared small target images.  ...  However, existing deep learning methods are based on the global view to extract features from infrared images and ignore the imaging locality of small targets.  ... 
arXiv:2108.06054v3 fatcat:ttwgwconpjbmhnsu44lqnhj7bq

Fisheye-Based Smart Control System for Autonomous UAV Operation

Donggeun Oh, Junghee Han
2020 Sensors  
Especially, many ML-based (Machine-Learning-based) methods have been proposed for autonomous path finding.  ...  Specifically, we devised Multi-Layer HVIN (Hierarchical VIN) methods that increase the area applicable to autonomous flight by overlaying multiple layers.  ...  Multi-Layer HVIN. Fisheye HVIN Multi-Layer HVIN CPU Usage Planning is performed only in two layers, local and global.  ... 
doi:10.3390/s20247321 pmid:33419238 fatcat:uufpp7twb5c5tdfuc2ouie6bvi

Active learning for relation type extension with local and global data views

Ang Sun, Ralph Grishman
2012 Proceedings of the 21st ACM international conference on Information and knowledge management - CIKM '12  
We present LGCo-Testing, an active learning system for relation type extension based on local and global views of relation instances.  ...  Locally, we extract features from the sentence that contains the instance. Globally, we measure the distributional similarity between instances from a 2 billion token corpus.  ...  Conclusion We have presented LGCo-Testing, a multi-view active learning approach for relation type extension based on local and global views.  ... 
doi:10.1145/2396761.2398409 dblp:conf/cikm/SunG12 fatcat:rxy72pbfbnhipkeetqjfauys6i

Transferable Attention for Domain Adaptation

Ximei Wang, Liang Li, Weirui Ye, Mingsheng Long, Jianmin Wang
2019 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Existing methods of adversarial domain adaptation mainly align the global images across the source and target domains.  ...  We implement two types of complementary transferable attention: transferable local attention generated by multiple region-level domain discriminators to highlight transferable regions, and transferable  ...  Acknowledgements This work was supported by the National Key R&D Program of China (No. 2016YFB1000701) and Natural Science Foundation of China (61772299, 61502265, 71690231).  ... 
doi:10.1609/aaai.v33i01.33015345 fatcat:cfm2vfhnaneslo2yckn5c762ta
« Previous Showing results 1 — 15 out of 178,631 results