31,223 Hits in 6.7 sec

Post Training in Deep Learning with Last Kernel [article]

Thomas Moreau, Julien Audiffren
2017 arXiv   pre-print
In this article, we propose an extra training step, called post-training, which only optimizes the last layer of the network.  ...  One of the main challenges of deep learning methods is the choice of an appropriate training strategy.  ...  In this setting, learning the weights of the last layer corresponds to learning the weights for the kernel associated to the feature map given by the previous layers.  ... 
arXiv:1611.04499v2 fatcat:b7fynkplnzg3heqjihymq5jf6a

Deep Learning for identifying radiogenomic associations in breast cancer [article]

Zhe Zhu, Ehab Albadawy, Ashirbani Saha, Jun Zhang, Michael R. Harowicz, Maciej A. Mazurowski
2017 arXiv   pre-print
Conclusion: Deep learning may play a role in discovering radiogenomic associations in breast cancer.  ...  Three different deep learning approaches were used to classify the tumor according to their molecular subtypes: learning from scratch where only tumor patches were used for training, transfer learning  ...  Methods Overview Three different deep learning approaches were used in our experiment: training from scratch, transfer learning and offthe-shelf deep features approach.  ... 
arXiv:1711.11097v1 fatcat:jfp337nh4zd4nayaaac2ndizri

Deep learning analysis of breast MRIs for prediction of occult invasive disease in ductal carcinoma in situ [article]

Zhe Zhu, Michael Harowicz, Jun Zhang, Ashirbani Saha, Lars J. Grimm, E.Shelley Hwang, Maciej A. Mazurowski
2017 arXiv   pre-print
In the first approach, we adopted the transfer learning strategy, in which a network pre-trained on a large dataset of natural images is fine-tuned with our DCIS images.  ...  Results: The best classification performance was obtained using the deep features approach with GoogleNet model pre-trained on ImageNet as the feature extractor and a polynomial kernel SVM used as the  ...  DEEP LEARNING PLATFORM We used deep learning in two ways: transfer learning and off-the-shelf deep features.  ... 
arXiv:1711.10577v1 fatcat:4cw3qzj2arhwzcln7oxz6ffe3i

Last Layer Marginal Likelihood for Invariance Learning [article]

Pola Schwöbel, Martin Jørgensen, Sebastian W. Ober, Mark van der Wilk
2022 arXiv   pre-print
The Bayesian paradigm for model selection provides a path towards end-to-end learning of invariances using only the training data, by optimising the marginal likelihood.  ...  Our results indicate that once more sophisticated approximations become available the marginal likelihood is a promising approach for invariance learning in neural networks.  ...  In our setting, i.e. when trying to combine deep kernel learning with invariance learning, joint training produces overfit weights which results in simplistic features with little intra-class variation  ... 
arXiv:2106.07512v2 fatcat:5ojyoa62kra6jlw4hoaqnwqema

DeepCorn: A Semi-Supervised Deep Learning Method for High-Throughput Image-Based Corn Kernel Counting and Yield Estimation [article]

Saeed Khaki, Hieu Pham, Ye Han, Andy Kuhl, Wade Kent, Lizhi Wang
2020 bioRxiv   pre-print
Recent advances in machine learning, in particular deep learning, have shown promise in mitigating this bottleneck.  ...  In this paper, we propose a novel deep learning method for counting on-ear corn kernels in-field to aid in the gathering of real-time data and, ultimately, to improve decision making to maximize yield.  ...  Although traditional machine learning approaches have seen success in agriculture, the current state of the art in HTP is with the application of deep learning.  ... 
doi:10.1101/2020.11.09.375535 fatcat:f5qtlj4qffbuljsp2fixgiahp4

The Alexnet-ResNet-Inception Network for Classifying Fruit Images [article]

liu wenzhong
2020 bioRxiv   pre-print
In this study, the Alexnet, ResNet, and Inception networks were integrated to construct a deep convolutional neural network named Interfruit, which was then utilized in identifying various types of fruit  ...  To sum up, findings in this study indicate that the classification system Interfruitr ecognizes fruits with high accuracy and has a broad application prospect.  ...  Availability of data and materials All data sets and codes used in this study are available at  ... 
doi:10.1101/2020.02.09.941039 fatcat:wep2o3ojhrbnfmktcyxkb4rgxe

Classification of Discussion Threads in MOOC Forums Based on Deep Learning

Lin FENG, Lei WANG, Sheng-lan LIU, Guo-chao LIU
2018 DEStech Transactions on Computer Science and Engineering  
In this paper, we propose a model constructed by the convolutional neural network in deep learning to classify discussion threads in MOOC forums.  ...  As the only way for students and instructors to communicate in Massive Open Online Course (MOOCs), MOOC forum plays an important role in supervision of students' learning.  ...  In recent years, deep learning is a hot point in the field of machine learning.  ... 
doi:10.12783/dtcse/wcne2017/19907 fatcat:itwye2ntzncsrb6xc4jw56czdm

Densely Connected Large Kernel Convolutional Network for Semantic Membrane Segmentation in Microscopy Images

Dongnan Liu, Donghao Zhang, Siqi Liu, Yang Song, Haozhe Jia, Dagan Feng, Yong Xia, Weidong Cai
2018 2018 25th IEEE International Conference on Image Processing (ICIP)  
In our work, we propose a network with a ResNet encoder and densely connected decoder with large kernels, and then refinement with simple morphological post-possessing.  ...  Semantic segmentation of neurons thus becomes an important technique in bioinformatics. Deep learning approaches have shown promising performance in various semantic segmentation problems.  ...  As our model is based on the deep neural network, we choose to perform a more detailed comparison with approaches based on deep learning models.  ... 
doi:10.1109/icip.2018.8451775 dblp:conf/icip/LiuZLSJFXC18 fatcat:y6kl6wcumrf4fapztpgd7ocffe

Understanding sequence conservation with deep learning [article]

Yi Li, Daniel Quang, Xiaohui Xie
2017 bioRxiv   pre-print
DeepCons utilizes hundreds of convolution kernels to detect features within DNA sequences, and automatically learns these kernels after training the CNN model using 887,577 conserved elements and a similar  ...  Results: We present a deep learning framework, called DeepCons, to uncover potential functional elements within conserved sequences.  ...  Recent advances in deep learning, specifically in solving sequence-based problems in genomics with convolutional neural networks [3, 4, 5, 6] , provide a new powerful method to study sequence conservation  ... 
doi:10.1101/103929 fatcat:s4o3npazqjfcjnlvyblc56u3hy

Convolutional Neural Networks for Multimedia Sentiment Analysis [chapter]

Guoyong Cai, Binbin Xia
2015 Lecture Notes in Computer Science  
Two individual CNN architectures are used for learning textual features and visual features, which can be combined as input of another CNN architecture for exploiting the internal relation between text  ...  Experimental Setup and Results Datasets In our work, training dataset is constructed with randomly chosen 20K image posts (one image post consists of one image and corresponding description) from SentiBank  ...  In the field of natural language processing (NLP), works with deep learning methods were also widely used. As a challenging task NLP, sentiment analysis has been studied in various ways.  ... 
doi:10.1007/978-3-319-25207-0_14 fatcat:qtim3nkrmnegpjzsxl273ligie


L. Ding, H. Li, C. Hu, W. Zhang, S. Wang
2018 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
Taking the Google image with 0.3 meter spatial resolution in Ludian area of Yunnan Province as an example, the image segmentation object was taken as the basic unit, and the pre-trained AlexNet deep convolution  ...  And the spectral features, AlexNet features and GLCM texture features are combined with multi-kernel learning and SVM classifier, finally the classification results were compared and analyzed.  ...  In order to analyze the effectiveness of the proposed method, the same training samples and testing samples were used to compare 3 methods of the experiments. (1) Method 1:Multi-kernel learning and SVM  ... 
doi:10.5194/isprs-archives-xlii-3-277-2018 fatcat:rbt5rorhzrhx5clg6wi7fofoau

ReStoCNet: Residual Stochastic Binary Convolutional Spiking Neural Network for Memory-Efficient Neuromorphic Computing [article]

Gopalakrishnan Srinivasan, Kaushik Roy
2019 arXiv   pre-print
In addition, we introduce residual connections between the stacked convolutional layers to improve the hierarchical feature learning capability of deep SNNs.  ...  binary kernels forming ReStoCNet in a layer-wise unsupervised manner.  ...  In this work, we use fully-connected layer of ReLU neurons trained with backpropagation algorithm commonly used for deep learning networks since we are primarily interested in evaluating the efficacy of  ... 
arXiv:1902.04161v1 fatcat:y4hutkwm35ambbmiqepg55bdqa

Understanding Sequence Conservation With Deep Learning

Yi Li, Daniel Quang, Xiaohui Xie
2017 Proceedings of the 8th ACM International Conference on Bioinformatics, Computational Biology,and Health Informatics - ACM-BCB '17  
Recent advances in deep learning, specifically in solving sequence-based problems in genomics with convolutional neural networks [3, 4, 5, 6] , provide a new powerful method to study sequence conservation  ...  In this paper we present a deep learning method for studying sequence conservation (DeepCons).  ...  In summary, we have developed a new deep learning framework for studying conserved sequences using convolutional neural networks.  ... 
doi:10.1145/3107411.3107425 dblp:conf/bcb/LiQX17 fatcat:gyub746ntncldmfyuet3rw3eiu

Enabling Spike-based Backpropagation in State-of-the-art Deep Neural Network Architectures [article]

Chankyu Lee, Syed Shakib Sarwar, Kaushik Roy
2019 arXiv   pre-print
MNIST, SVHN and CIFAR-10 datasets compared to other SNNs trained with spike-based learning.  ...  Our experiments show the effectiveness of the proposed spike-based learning strategy on state-of-the-art deep networks (VGG and Residual architectures) by achieving the best classification accuracies in  ...  Acknowledgement This work was supported in part by C-BRIC, one of six centers in JUMP, a Semiconductor Research Corporation (SRC) program sponsored by DARPA, the National Science Foundation, Intel Corporation  ... 
arXiv:1903.06379v3 fatcat:6ffydepbffda5c7ofskxvzslay

Deep Learning for PET Image Reconstruction

Andrew J. Reader, Guillaume Corda, Abolfazl Mehranian, Casper da Costa-Luis, Sam Ellis, Julia A. Schnabel
2020 IEEE Transactions on Radiation and Plasma Medical Sciences  
In contrast, model-based or physics-informed deep-learning uses existing advances in PET image reconstruction, replacing conventional components with deep-learning data-driven alternatives, such as for  ...  The direct deep-learning methodology is then reviewed in the context of PET reconstruction.  ...  Deep Learning for Preprocessing and Post-Processing As mentioned, deep learning for post-reconstruction processing (or even for preprocessing of the raw sinogram data), is not under consideration in this  ... 
doi:10.1109/trpms.2020.3014786 fatcat:rxj5idat3fhfdbl2oeqe3hw5na
« Previous Showing results 1 — 15 out of 31,223 results