Filters








126,643 Hits in 3.5 sec

DeepForge: A Machine Learning Gateway for Scientific Workflow Design

Peter Volgyesi, Akos Ledeczi, Brian Broll, Tamas Budavari
2018 Figshare  
DeepForge is an open source platform for deep learning designed for promoting reproducibility, simplicity and rapid development within diverse scientific domains.  ...  provided for working with neural networks: Architectures and Layers Editing a training operation with net and criterion references and a data input, trainset Viewing Formal specification of the language  ...  Model-Integrated Computing • The process of using domain specific abstractions for developing systems or applications • The domain specific model (DSM) is at the center of the workflow • Developed to aid  ... 
doi:10.6084/m9.figshare.6170933.v1 fatcat:aagjpwm7ujhgpnkdhbsb6ykmwq

DeepForge: A Machine Learning Gateway for Scientific Workflow Design

Peter Volgyesi, Akos Ledeczi, Brian Broll, Tamas Budavari
2018 Figshare  
DeepForge is an open source platform for deep learning designed for promoting reproducibility, simplicity and rapid development within diverse scientific domains.  ...  provided for working with neural networks: Architectures and Layers Editing a training operation with net and criterion references and a data input, trainset Viewing Formal specification of the language  ...  Model-Integrated Computing • The process of using domain specific abstractions for developing systems or applications • The domain specific model (DSM) is at the center of the workflow • Developed to aid  ... 
doi:10.6084/m9.figshare.6170933 fatcat:by6pjrmzejgkdk7xqogytj3wzq

Transfer Learning with Binary Neural Networks [article]

Sam Leroux, Steven Bohez, Tim Verbelen, Bert Vankeirsbilck, Pieter Simoens, Bart Dhoedt
2017 arXiv   pre-print
have dedicated accelerators for deep neural networks.  ...  We propose a transfer learning based architecture where we first train a binary network on Imagenet and then retrain part of the network for different tasks while keeping most of the network fixed.  ...  Steven Bohez is funded by a Ph.D. grant of the Agency for Innovation by Science and Technology in Flanders (IWT).  ... 
arXiv:1711.10761v1 fatcat:fkyzh5ov6zcrxka3ykujfa5h5m

DeepForge: A Scientific Gateway for Deep Learning

Brian Broll, Miklos Maroti, Peter Volgyesi, Akos Ledeczi
2018 Figshare  
We introduce DeepForge, a gateway to deep learning for the scientific community. DeepForge is designed to lower the barrier to entry and facilitate the rapid development of deep learning models.  ...  platform for the scientific community.  ...  Accessibility DeepForge uses WebGME, a framework for creating domain specific modeling environments, to create a domain specific modeling language for neural networks and the concepts described in Section  ... 
doi:10.6084/m9.figshare.7092272.v1 fatcat:2fbm6gji4vbzbctis7em6bzqti

spectrai: A deep learning framework for spectral data [article]

Conor C. Horgan, Mads S. Bergholt
2021 arXiv   pre-print
However, the application of deep learning to spectral data remains a complex task due to the need for augmentation routines, specific architectures for spectral data, and significant memory requirements  ...  Here we present spectrai, an open-source deep learning framework designed to facilitate the training of neural networks on spectral data and enable comparison between different methods.  ...  However, the successful application of deep learning to different applications requires significant expertise, with task-specific design decisions for network architecture, loss function, learning rate  ... 
arXiv:2108.07595v1 fatcat:gddhfx342rhgzjgaxeljgl5qxe

Modular Mechanistic Networks: On Bridging Mechanistic and Phenomenological Models with Deep Neural Networks in Natural Language Processing [article]

Simon Dobnik, John D. Kelleher
2019 arXiv   pre-print
Examining some recent approaches in deep learning we argue that deep neural networks incorporate both perspectives and, furthermore, that leveraging this aspect of deep learning may help in solving complex  ...  problems within language technology, such as modelling language and perception in the domain of spatial cognition.  ...  Moreover, we highlight that many of the recent advances in deep learning for NLP are not based on unconstrained neural networks but rather that these networks have task specific architectures that encode  ... 
arXiv:1807.09844v2 fatcat:enb5w34frbaqpeglk2rmtw3rja

Deep learning—Accelerating Next Generation Performance Analysis Systems?

Heike Brock
2018 Proceedings (MDPI)  
Deep neural network architectures show superior performance in recognition and prediction tasks of the image, speech and natural language domains.  ...  In particular, it discusses aspects of data acquisition, processing and network modeling. Analysis suggests the advantage of deep neural networks under difficult and noisy data conditions.  ...  System and Data Requirements As discussed in Section 2, deep neural network architectures are able to learn subtle connections within a set of training data without domain knowledge and information loss  ... 
doi:10.3390/proceedings2060303 fatcat:oyu77gvw2zadpfft5v7q5fzumq

2019 Index IEEE Journal on Emerging and Selected Topics in Circuits and Systems Vol. 9

2019 IEEE Journal on Emerging and Selected Topics in Circuits and Systems  
Hanhart, P., +, JETCAS March 2019 71-83 CNNP-v2: A Memory-Centric Architecture for Low-Power CNN Processor on Domain-Specific Mobile Devices.  ...  ., +, JETCAS June 2019 398-410 Energy efficiency CNNP-v2: A Memory-Centric Architecture for Low-Power CNN Processor on Domain-Specific Mobile Devices.  ...  P Parallel architectures Hardware Design of a Context-Preserving Filter-Reorganized CNN for Super-Resolution. Lee  ... 
doi:10.1109/jetcas.2019.2958462 fatcat:faydtl5ymjfcxdc2nfrkgg7nxi

Brain Tumor Segmentation in MRI Images

Adarsh Dhiman, Prof. B.S. Satpute
2019 International Journal of Research in Advent Technology  
In this paper, we recommend using U-Net based deep convolutional neural network to carry out brain tumor segmentation.  ...  Before the advent of the deep learning applications, a specialist in the field of medical domain used to construct the features and it demanded the special knowledge in the medical field.  ...  LITERATURE REVIEW A. U-Net based deep Convolutional Neural Networks for Brain Tumor Segmentation The skip architecture is intended to manage variable edge issue.  ... 
doi:10.32622/ijrat.78201916 fatcat:2nlimi6qzragrir6rdl36fx3du

Matters of Neural Network Repository Designing for Analyzing and Predicting of Spatial Processes

Stanislav A. Yamashkin, Anatoliy A. Yamashkin, Ekaterina O. Yamashkina, Anastasiya A. Kamaeva
2021 International Journal of Advanced Computer Science and Applications  
An ontological model of a deep neural network repository for spatial data analysis is decomposed into the domain of deep machine learning models, problems being solved and data.  ...  The issues of architecture development and software implementation of a repository of deep neural network models for spatial data analysis are considered, based on a new ontological model, which makes  ...  to deep neural network models. 6) Development of interfaces for obtaining structured information about specific neural network models. 7) Development of a subsystem for visualizing deep machine learning  ... 
doi:10.14569/ijacsa.2021.0120503 fatcat:jyvlustezvcdphgfyb4f3xgeji

Character Level based Detection of DGA Domain Names

Bin Yu, Jie Pan, Jiaming Hu, Anderson Nascimento, Martine De Cock
2018 2018 International Joint Conference on Neural Networks (IJCNN)  
Training and evaluating on a dataset with 2M domain names shows that there is surprisingly little difference between various convolutional neural network (CNN) and recurrent neural network (RNN) based  ...  Recently several different deep learning architectures have been proposed that take a string of characters as the raw input signal and automatically derive features for text classification.  ...  Independent of the work on deep networks for DGA detection, other deep learning approaches for character based text classification have recently been proposed, including deep neural network architectures  ... 
doi:10.1109/ijcnn.2018.8489147 dblp:conf/ijcnn/YuPHNC18 fatcat:hd3ztzvd75crtpaqzo2ij4hpla

A Machine Learning Gateway for Scientific Workflow Design

Brian Broll, Umesh Timalsina, Péter Völgyesi, Tamás Budavári, Ákos Lédeczi, Manuel E. Acacio Sanchez
2020 Scientific Programming  
The paper introduces DeepForge, a gateway to deep learning for scientific computing.  ...  The tool currently supports TensorFlow/Keras, but its extensible architecture enables easy integration of additional platforms.  ...  Model Architecture. is example uses a deep-convolutional neural network similar to the architecture used in [21] . e network starts with a convolution layer that takes an input image of size 64 × 64 with  ... 
doi:10.1155/2020/8867380 fatcat:bzbyuieggbgr7kqlfn6os5leem

Transfer learning-based Plant Disease Detection

2021 International journal for innovative engineering and management research  
Deep Neural Networks in the field of Machine Learning (ML) are broadly used for deep learning.  ...  In this paper, we have shown the use of deep neural networks for plant disease detection, through image classification.  ...  Deep Convolutional Neural Networks A Convolutional Neural Network (CNN) could be a stack of non-linear transformation functions that are learned from data.  ... 
doi:10.48047/ijiemr/v10/i03/99 fatcat:7gwa3st6szdqtbcl5kp26vqbje

Network Architecture Search for Domain Adaptation [article]

Yichen Li, Xingchao Peng
2020 arXiv   pre-print
In this paper, we present Neural Architecture Search for Domain Adaptation (NASDA), a principle framework that leverages differentiable neural architecture search to derive the optimal network architecture  ...  Deep networks have been used to learn transferable representations for domain adaptation.  ...  in deriving an optimal neural architecture specific for domain transfer.  ... 
arXiv:2008.05706v1 fatcat:fjjiakdbhnf43fnxc7jxx7u5zu

A multi domains short message sentiment classification using hybrid neural network architecture

Devi Munandar, Andri Fachrur Rozie, Andria Arisal
2021 Bulletin of Electrical Engineering and Informatics  
This paper explores several deep learning methods, such as multilayer perceptron (MLP), convolutional neural network (CNN), long short-term memory (LSTM), and builds combinations of those three architectures  ...  Those advantages are useful for different domain datasets.  ...  Ivonesia Solusi Data (Ivosight) which have provided us with a very valuable labeled dataset. All authors contributed equally as main contributors of this paper, read, and approved the final paper.  ... 
doi:10.11591/eei.v10i4.2790 fatcat:44w5hti6svdqzppoxgx4vlmpza
« Previous Showing results 1 — 15 out of 126,643 results