Filters








255,426 Hits in 5.1 sec

Towards a Theory of Extractable Functions [chapter]

Ran Canetti, Ronny Ramzi Dakdouk
2009 Lecture Notes in Computer Science  
Examples include reductions from extractable weak one-way functions to extractable strong ones, from extractable pseudorandom generators to extractable pseudorandom functions, and from extractable one-way  ...  Extractability of functions was defined by the authors (ICALP'08) in the context of perfectly one-way functions.  ...  Any deterministic one-way function is not even weakly extractable. That is, any deterministic one-way function satisfies statement 1 of Theorem 1.  ... 
doi:10.1007/978-3-642-00457-5_35 fatcat:zgu5ebozurdqvanv4er47a3sfq

Weakly Supervised Cascaded Convolutional Networks [article]

Ali Diba, Vivek Sharma, Ali Pazandeh, Hamed Pirsiavash, Luc Van Gool
2016 arXiv   pre-print
The final stage of both architectures is a part of a convolutional neural network that performs multiple instance learning on proposals extracted in the previous stage(s).  ...  Our experiments on the PASCAL VOC 2007, 2010, 2012 and large scale object datasets, ILSVRC 2013, 2014 datasets show improvements in the areas of weakly-supervised object detection, classification and localization  ...  To improve the detection performance, object proposal generation, feature extraction, and MIL are trained in a cascaded manner, in an end-to-end way. We propose two architectures.  ... 
arXiv:1611.08258v1 fatcat:cbdt7v5m2jdr7bypq4hvec23ru

Weakly-supervised Graph Meta-learning for Few-shot Node Classification [article]

Kaize Ding, Jianling Wang, Jundong Li, James Caverlee, Huan Liu
2021 arXiv   pre-print
Based on a new robustness-enhanced episodic training, Meta-GHN is meta-learned to hallucinate clean node representations from weakly-labeled data and extracts highly transferable meta-knowledge, which  ...  Extensive experiments demonstrate the superiority of Meta-GHN over existing graph meta-learning studies on the task of weakly-supervised few-shot node classification.  ...  By minimizing the above loss function, GHN is able to learn a generic classifier for a specific -way -set meta-training task and further extract highly transferrable metaknowledge from weakly-labeled data  ... 
arXiv:2106.06873v1 fatcat:5k3jesbd5nbglnncehbv6xu4m4

Aircraft Image Recognition Network Based on Hybrid Attention Mechanism

Yanfeng Wang, Yinan Chen, Runmin Liu, Tongguang Ni
2022 Computational Intelligence and Neuroscience  
Using two-channel ResNet-34 as a characteristic extraction function, the depth of network is increased to improve fine-grained characteristic extraction capability without increasing the output characteristic  ...  the recognition precision rate of the original model is improved effectively by using this method, and the recognition precision rate is higher than most of the existing mainstream aircraft recognition ways  ...  BA-CNN consists of two ResNet-34 networks as the characteristic extraction function and adds the channel attention module and spatial attention module to the two-way characteristic function, and the convolutional  ... 
doi:10.1155/2022/4189500 pmid:35479608 pmcid:PMC9038415 fatcat:66jgedmsczgadaqmn7dyrvlrxu

A Variational Approach to Weakly Supervised Document-Level Multi-Aspect Sentiment Classification

Ziqian Zeng, Wenxuan Zhou, Xin Liu, Yangqiu Song
2019 Proceedings of the 2019 Conference of the North  
We show that our method can outperform weakly supervised baselines on TripAdvisor and BeerAdvocate datasets and can be comparable to the state-of-the-art supervised method with hundreds of labels per aspect  ...  These word pairs can be extracted by using dependency parsers and simple rules.  ...  DMSC of unsupervised, weakly supervised, and supervised methods on TripAdvisor and BeerAdvocate.  ... 
doi:10.18653/v1/n19-1036 dblp:conf/naacl/ZengZLS19 fatcat:4v75kaoj2janrmounn57vy4z3a

A Variational Approach to Weakly Supervised Document-Level Multi-Aspect Sentiment Classification [article]

Ziqian Zeng, Wenxuan Zhou, Xin Liu, Yangqiu Song
2019 arXiv   pre-print
We show that our method can outperform weakly supervised baselines on TripAdvisor and BeerAdvocate datasets and can be comparable to the state-of-the-art supervised method with hundreds of labels per aspect  ...  These word pairs can be extracted by using dependency parsers and simple rules.  ...  For weakly supervised methods, LRR is the only open-source baseline in the literature on weakly supervised DMSC, and our method outperforms LRR by 6% and 16% on TripAdvisor and BeerAdvocate datasets.  ... 
arXiv:1904.05055v1 fatcat:4snwpguqtnhv5oyt4petqnavzq

Efficient Labelling of Pedestrian Supervisions

Kyaw Kyaw Htike
2016 ELCVIA Electronic Letters on Computer Vision and Image Analysis  
By evaluating on publicly available datasets, we demonstrate that our weakly supervised algorithm reduces the cost of manual annotation of pedestrians by more than four times while achieving similar performance  ...  The algorithm makes use of a pedestrian prior learnt in an unsupervised way from the input video and this prior is fused with the given weak supervision information in a systematic manner.  ...  One such way to calibrate it by using Platt scaling [40] .  ... 
doi:10.5565/rev/elcvia.881 fatcat:wmori6q6yzbxvo4cmocuww76py

Bilinear CNN Model for Fine-Grained Classification Based on Subcategory-Similarity Measurement

Xinghua Dai, Shengrong Gong, Shan Zhong, Zongming Bao
2019 Applied Sciences  
weakly supervised classification models.  ...  To further improve classification accuracy, the loss function is designed by weighting triplet loss and softmax loss.  ...  They are both based on the widely used VGG-16, and the architectures of VGG-16 are modified to accommodate different functions: Before using B-CNN to extract features, a weakly supervised localization  ... 
doi:10.3390/app9020301 fatcat:vnotxlzd7zekha4yq2iooealje

A Black-Box Construction of a CCA2 Encryption Scheme from a Plaintext Aware (sPA1) Encryption Scheme [chapter]

Dana Dachman-Soled
2014 Lecture Notes in Computer Science  
Our work extends and improves on this result by showing that a full CCA2-secure encryption scheme can be constructed from the same assumptions. 1 A CCA1-encryption scheme is one where the adversary has  ...  We present a construction of a CCA2-secure encryption scheme from a plaintext aware (sPA1), weakly simulatable public key encryption scheme.  ...  Such schemes can be constructed in a black-box way from one-way functions [25, 30] , and thus from any semanticallysecure encryption scheme (Gen, Enc, Dec).  ... 
doi:10.1007/978-3-642-54631-0_3 fatcat:44fbm6a67jbijpu6532midiy7q

W-TALC: Weakly-supervised Temporal Activity Localization and Classification [article]

Sujoy Paul, Sourya Roy, Amit K Roy-Chowdhury
2018 arXiv   pre-print
Recent years have witnessed a substantial influx of tagged videos on the Internet, which can serve as a rich source of weakly-supervised training data.  ...  The proposed network can be divided into two sub-networks, namely the Two-Stream based feature extractor network and a weakly-supervised module, which we learn by optimizing two complimentary loss functions  ...  Note that l i is not only dependent on the video index i, but also on the feature extraction procedure used. These matrices become the input to our weakly-supervised learning module.  ... 
arXiv:1807.10418v3 fatcat:k4okqt7ti5axrcy4spdsanfxty

Weakly Supervised Cascaded Convolutional Networks

Ali Diba, Vivek Sharma, Ali Pazandeh, Hamed Pirsiavash, Luc Van Gool
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
The final stage of both architectures is a part of a convolutional neural network that performs multiple instance learning on proposals extracted in the previous stage(s).  ...  The first stage of both architectures extracts best candidate of class specific region proposals by training a fully convolutional network.  ...  To improve the detection performance, object proposal generation, feature extraction, and MIL are trained in a cascaded manner, in an end-to-end way. We propose two architectures.  ... 
doi:10.1109/cvpr.2017.545 dblp:conf/cvpr/DibaSPPG17 fatcat:btei6ijyxbf4xc6x4hahnirs7u

Assigning function to protein-protein interactions: a weakly supervised BioBERT based approach using PubMed abstracts [article]

Aparna Elangovan, Melissa Davis, Karin Verspoor
2022 arXiv   pre-print
Here, we aim to label the function type of PPIs by extracting relationships described in PubMed abstracts.  ...  Method: We create a weakly supervised dataset from the IntAct PPI database containing interacting protein pairs with annotated function and associated abstracts from the PubMed database.  ...  Probability thresholding is one way to improve the quality; the higher the threshold the more likely that the predictions are of high quality.  ... 
arXiv:2008.08727v3 fatcat:bs6dc447bvhcbnfhu32wpfmqjy

Weakly Supervised Video Anomaly Detection Based on 3D Convolution and LSTM

Zhen Ma, José J. J. Machado, João Manuel R. S. Tavares
2021 Sensors  
Weakly supervised video anomaly detection is a recent focus of computer vision research thanks to the availability of large-scale weakly supervised video datasets.  ...  However, most existing research works are limited to the frame-level classification with emphasis on finding the presence of specific objects or activities.  ...  Based on the way of human intervention and the information used for training, usual deep learning methods for video anomaly detection can be classified into supervised, unsupervised, and weakly supervised  ... 
doi:10.3390/s21227508 pmid:34833584 pmcid:PMC8620488 fatcat:4k4qjqc6krctnhjp7qw6jh3xxy

EXTRACTION OF URANIUM FROM SEA WATER BY MEANS OF FIBROUS COMPLEX ADSORBENTS

Tokuhisa Miyarnatsu, Noboru Oguchi, Yoshihiko Kanchiku, Takanobu Aovagi
1982 Sen'i Gakkaishi  
Among the adsorbents examined, the following ones demon strated excellent properties for the recovery of uranium from sea water. a) A fibrous, weakly acidic cation exchanger was treated with titanyl sulfate  ...  Fibrous complex adsorbents for uranium extraction from sea water were prepared by introducing titanic acid or basic zinc carbonate as effective constituents into fibrous ion exchangers.  ...  On the surface of QCS-Zn and AC-Ti precipitates can be observed, but on the another adsorbents there are no precipitates. Fig. 3 . 3 Titration curve of HPPA having group as functional group.  ... 
doi:10.2115/fiber.38.12_t537 fatcat:isn22pm2s5devcyibbgcbj7j74

Road Extraction from Very High Resolution Images Using Weakly labeled OpenStreetMap Centerline

Songbing Wu, Chun Du, Hao Chen, Yingxiao Xu, Ning Guo, Ning Jing
2019 ISPRS International Journal of Geo-Information  
Due to the success of semantic segmentation based on deep learning in the application of computer vision, extracting road networks from VHR (Very High Resolution) imagery becomes a method of updating geographic  ...  In this paper, we propose a novel weakly supervised method to extract road networks from VHR images using only the OSM (OpenStreetMap) road centerline as training data instead of high quality pixel-wise  ...  Evaluation Road Extraction Using Normalized Cut Loss Combined With Partial Losses We examined the overall road extraction approach based on the weakly supervised centerline annotation.  ... 
doi:10.3390/ijgi8110478 fatcat:ji6wqnk4arh3liugukvz6x3sqa
« Previous Showing results 1 — 15 out of 255,426 results