Filters








153,580 Hits in 3.8 sec

Adaptive Selection of Deep Learning Models on Embedded Systems [article]

Ben Taylor, Vicent Sanz Marco, Willy Wolff, Yehia Elkhatib, Zheng Wang
2018 arXiv   pre-print
The recent ground-breaking advances in deep learning networks ( DNNs ) make them attractive for embedded systems.  ...  We apply our approach to the image classification task and evaluate it on a Jetson TX2 embedded deep learning platform using the ImageNet ILSVRC 2012 validation dataset.  ...  CONCLUSION This paper has presented a novel scheme to dynamically select a deep learning model to use on an embedded device.  ... 
arXiv:1805.04252v1 fatcat:bakdwt5xvjd3vjsop4g45t7lbe

Adaptive deep learning model selection on embedded systems

Ben Taylor, Vicent Sanz Marco, Willy Wolff, Yehia Elkhatib, Zheng Wang
2018 Proceedings of the 19th ACM SIGPLAN/SIGBED International Conference on Languages, Compilers, and Tools for Embedded Systems - LCTES 2018  
The recent ground-breaking advances in deep learning networks (DNNs) make them attractive for embedded systems.  ...  We apply our approach to the image classification task and evaluate it on a Jetson TX2 embedded deep learning platform using the ImageNet ILSVRC 2012 validation dataset.  ...  CONCLUSION This paper has presented an adaptive scheme to dynamically select a deep learning model to use on an embedded device.  ... 
doi:10.1145/3211332.3211336 dblp:conf/lctrts/TaylorMWE018 fatcat:ypn5zmna7vbwfmaxrfukbdzbxi

An Embedded System for Image-based Crack Detection by using Fine-Tuning model of Adaptive Structural Learning of Deep Belief Network [article]

Shin Kamada, Takumi Ichimura
2021 arXiv   pre-print
In our research, an adaptive structural learning method of Restricted Boltzmann Machine (Adaptive RBM) and Deep Belief Network (Adaptive DBN) have been developed as a deep learning model.  ...  Deep learning has been a successful model which can effectively represent several features of input space and remarkably improve image recognition performance on the deep architectures.  ...  The NVIDIA Jetson series are known to be a tiny embedded system which enables fast inference of deep learning.  ... 
arXiv:2110.13145v1 fatcat:k7ijosnnlrcjvk5g3icky4rn4i

Neural Network based Deep Transfer Learning for Cross-domain Dependency Parsing [article]

Zhentao Xia, Likai Wang, Weiguang Qu, Junsheng Zhou, Yanhui Gu
2019 arXiv   pre-print
In addition, to adapt three dif-ferent domains, we utilize neural network based deep transfer learning which transfers the pre-trained partial network in the source domain to be a part of deep neural network  ...  Our system is based on the stack-pointer networks(STACKPTR).  ...  parsing and the domain adaptation with deep transfer learning.  ... 
arXiv:1908.02895v1 fatcat:4cb2dmbgubfdhlcybqh22olvwi

Deep Visual-Semantic Quantization for Efficient Image Retrieval

Yue Cao, Mingsheng Long, Jianmin Wang, Shichen Liu
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
We propose Deep Visual-Semantic Quantization (DVSQ), which is the first approach to learning deep quantization models from labeled image data as well as the semantic information underlying general text  ...  The main contribution lies in jointly learning deep visual-semantic embeddings and visual-semantic quantizers using carefullydesigned hybrid networks and well-specified loss functions.  ...  Lab for Big Data System Software (NEL-BDSS), and Tsinghua National Lab for Information Science and Technology (TNList) Projects.  ... 
doi:10.1109/cvpr.2017.104 dblp:conf/cvpr/CaoL0L17 fatcat:jgzhlmcoeraqblejcdeoeovh6i

Machine Learning Approaches For Motor Learning: A Short Review [article]

Baptiste Caramiaux, Jules Françoise, Wanyu Liu, Téo Sanchez, Frédéric Bevilacqua
2020 arXiv   pre-print
We identify and describe three types of adaptation: Parameter adaptation in probabilistic models, Transfer and meta-learning in deep neural networks, and Planning adaptation by reinforcement learning.  ...  To conclude, we discuss challenges for applying these models in the domain of motor learning support systems.  ...  Conflict of Interest Statement The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest  ... 
arXiv:2002.04317v4 fatcat:q6ptmikhh5dd3iaizngkqoaqxq

Spectro-Temporal Deep Features for Disordered Speech Assessment and Recognition [article]

Mengzhe Geng, Shansong Liu, Jianwei Yu, Xurong Xie, Shoukang Hu, Zi Ye, Zengrui Jin, Xunying Liu, Helen Meng
2022 arXiv   pre-print
The final speaker adapted system using the proposed spectral basis embedding features gave an overall WER of 25.6% on the UASpeech test set of 16 dysarthric speakers  ...  Experiments conducted on the UASpeech corpus suggest the proposed spectro-temporal deep feature adapted systems consistently outperformed baseline i-Vector adaptation by up to 2.63% absolute (8.6% relative  ...  More recent researches investigated model adaptation of state-of-the-art deep neural network (DNN) based systems.  ... 
arXiv:2201.05554v1 fatcat:q32l6zo37ncprcsm2yxnpk52dm

Deep Meta-learning in Recommendation Systems: A Survey [article]

Chunyang Wang, Yanmin Zhu, Haobing Liu, Tianzi Zang, Jiadi Yu, Feilong Tang
2022 arXiv   pre-print
Recently, a growing number of studies on deep meta-learning based recommenddation systems have emerged for improving the performance under recommendation scenarios where available data is limited, e.g.  ...  However, since model training from scratch requires sufficient data, deep learning-based recommendation methods still face the bottlenecks of insufficient data and computational inefficiency.  ...  However, no previous survey centers on the deep meta-learning in recommendation systems.  ... 
arXiv:2206.04415v1 fatcat:w5rax6bjy5efjfmxunvf4j6kly

Neural Statistics for Click-Through Rate Prediction

Yanhua Huang, Hangyu Wang, Yiyun Miao, Ruiwen Xu, Lei Zhang, Weinan Zhang
2022 Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval  
Offline experiments on two public datasets validate the effectiveness of neural statistics against state-of-the-art models.  ...  With the success of deep learning, click-through rate (CTR) predictions are transitioning from shallow approaches to deep architectures.  ...  Instead of manually feeding crafted features into the model, neural statistics can be viewed as a deep innate prior, where the deep architecture learns statistical information internally.  ... 
doi:10.1145/3477495.3531762 fatcat:bpl7mc3umfasjfmmivczbnck6u

Automated Machine Learning for Deep Recommender Systems: A Survey [article]

Bo Chen, Xiangyu Zhao, Yejing Wang, Wenqi Fan, Huifeng Guo, Ruiming Tang
2022 arXiv   pre-print
Then we discuss the state-of-the-art AutoML approaches that automate the feature selection, feature embeddings, feature interactions, and system design in DRS.  ...  This article will give a comprehensive summary of automated machine learning (AutoML) for developing DRS models. We first provide an overview of AutoML for DRS models and the related techniques.  ...  First, this requires extensive expertise in deep learning and recommender systems.  ... 
arXiv:2204.01390v1 fatcat:ybiang7gajdkrljsrhbq6ih62m

Deep Music: Towards Musical Dialogue

Mason Bretan, Sageev Oore, Jesse Engel, Douglas Eck, Larry Heck
2017 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
We present a system that utilizes a deep autoencoder to learn semantic embeddings of musical input.  ...  Selection is based on a nearest neighbor search within the embedding space and for real-time application the search space is pruned using vector quantization.  ...  Interaction Adapting for performance The model is trained on a corpus of publicly available MIDI data that includes classical and jazz music.  ... 
doi:10.1609/aaai.v31i1.10544 fatcat:24yudlfq55h45ci34cnz7uj6sy

Prototypical Q Networks for Automatic Conversational Diagnosis and Few-Shot New Disease Adaption

Hongyin Luo, Shang-Wen Li, James Glass
2020 Interspeech 2020  
State-of-the-art dialog managers are usually driven by deep reinforcement learning models, such as deep Q networks (DQNs), which learn by interacting with a simulator to explore the entire action space  ...  The model calculates prototype embeddings with real conversations between doctors and patients, learning from them and simulator-augmented dialogs more efficiently.  ...  Since both many-and fewshot learning heavily depend on the quality of learned representations, these studies encouraged us to combine meta-learning and deep reinforcement learning models to improve the  ... 
doi:10.21437/interspeech.2020-1865 dblp:conf/interspeech/LuoLG20 fatcat:i4dpsbgrdndk3ft5acupstfgqa

Prototypical Q Networks for Automatic Conversational Diagnosis and Few-Shot New Disease Adaption [article]

Hongyin Luo, Shang-Wen Li, James Glass
2020 arXiv   pre-print
State-of-the-art dialog managers are usually driven by deep reinforcement learning models, such as deep Q networks (DQNs), which learn by interacting with a simulator to explore the entire action space  ...  The model calculates prototype embeddings with real conversations between doctors and patients, learning from them and simulator-augmented dialogs more efficiently.  ...  Since both many-and fewshot learning heavily depend on the quality of learned representations, these studies encouraged us to combine meta-learning and deep reinforcement learning models to improve the  ... 
arXiv:2005.11153v1 fatcat:anabw7ki2ba3pd3kpkfo2fj3g4

Machine Learning Approaches for Motor Learning: A Short Review

Baptiste Caramiaux, Jules Françoise, Wanyu Liu, Téo Sanchez, Frédéric Bevilacqua
2020 Frontiers in Computer Science  
We identify and describe three types of adaptation: Parameter adaptation in probabilistic models, Transfer and meta-learning in deep neural networks, and Planning adaptation by reinforcement learning.  ...  To conclude, we discuss challenges for applying these models in the domain of motor learning support systems.  ...  We then compiled the papers in a spreadsheet and conducted a selection based on the type of model adaptation, the modeling technique, the field, and the input data considered.  ... 
doi:10.3389/fcomp.2020.00016 fatcat:rfnjoptaa5bkbayvfcf6eniugq

Semi-supervised Adversarial Domain Adaptation for Seagrass Detection Using Multispectral Images in Coastal Areas

Kazi Aminul Islam, Victoria Hill, Blake Schaeffer, Richard Zimmerman, Jiang Li
2020 Data Science and Engineering  
However, a deep CNN model trained at one location usually does not generalize to other locations due to data distribution shifts.  ...  In this paper, we developed a semi-supervised domain adaptation method to generalize a trained deep CNN model to other locations for seagrass detection.  ...  This article has been reviewed by the Center for Environmental Measurement and Modeling and approved for publication.  ... 
doi:10.1007/s41019-020-00126-0 pmid:32685664 pmcid:PMC7357679 fatcat:ukinq74vnvcxvcfn7hyae7pcou
« Previous Showing results 1 — 15 out of 153,580 results