Filters








769 Hits in 5.3 sec

2020 Index IEEE Transactions on Knowledge and Data Engineering Vol. 32

2021 IEEE Transactions on Knowledge and Data Engineering  
., +, TKDE Aug. 2020 1502-1516 Feature Selection for Neural Networks Using Group Lasso Regularization. Feature Selective Projection with Low-Rank Embedding and Dual Laplacian Regularization.  ...  ., +, TKDE April 2020 659-673 Feature Selective Projection with Low-Rank Embedding and Dual Laplacian Regularization.  ... 
doi:10.1109/tkde.2020.3038549 fatcat:75f5fmdrpjcwrasjylewyivtmu

Graph Neural Networks: Methods, Applications, and Opportunities [article]

Lilapati Waikhom, Ripon Patgiri
2021 arXiv   pre-print
Various other domains conform to non-Euclidean space, for which graph is an ideal representation.  ...  In the last decade or so, we have witnessed deep learning reinvigorating the machine learning field.  ...  This enables deep embedding techniques based on Auto-encoders to embed local structural information inside the encoder, while the low embedding methods can simply not.  ... 
arXiv:2108.10733v2 fatcat:j3rfmkiwenebvmfyboasjmx4nu

Deep Learning in Protein Structural Modeling and Design [article]

Wenhao Gao, Sai Pooja Mahajan, Jeremias Sulam, Jeffrey J. Gray
2020 arXiv   pre-print
We dissect the emerging approaches using deep learning techniques for protein structural modeling, and discuss advances and challenges that must be addressed.  ...  We argue for the central importance of structure, following the "sequence -> structure -> function" paradigm.  ...  Variational auto-encoder (VAE) Auto-Encoders 50 (AEs), unlike the ones discussed so far, provide a model for unsupervised learning.  ... 
arXiv:2007.08383v1 fatcat:ynpdumcqnbel7duwffbork6s2u

MethylNet: an automated and modular deep learning approach for DNA methylation analysis

Joshua J. Levy, Alexander J. Titus, Curtis L. Petersen, Youdinghuan Chen, Lucas A. Salas, Brock C. Christensen
2020 BMC Bioinformatics  
DNAm deep learning approaches have not yet been formalized into user-friendly frameworks for execution, training, and interpreting models.  ...  Here, we describe MethylNet, a DNAm deep learning method that can construct embeddings, make predictions, generate new data, and uncover unknown heterogeneity with minimal user supervision.  ...  Consent for publication  ... 
doi:10.1186/s12859-020-3443-8 pmid:32183722 fatcat:njf2hraecfgyvmmuytebzbt23m

OperatorNet: Recovering 3D Shapes From Difference Operators [article]

Ruqi Huang and Marie-Julie Rakotosaona and Panos Achlioptas and Leonidas Guibas and Maks Ovsjanikov
2019 arXiv   pre-print
This paper proposes a learning-based framework for reconstructing 3D shapes from functional operators, compactly encoded as small-sized matrices.  ...  To this end we introduce a novel neural architecture, called OperatorNet, which takes as input a set of linear operators representing a shape and produces its 3D embedding.  ...  For example, several methods for shape interpolation have been proposed by designing deep neural networks, including auto-encoder architectures, and interpolating the latent vectors learned by such networks  ... 
arXiv:1904.10754v2 fatcat:pf6ufvp255fshjccfhflf7pe5q

Computational Analysis of Deformable Manifolds: from Geometric Modelling to Deep Learning [article]

Stefan C Schonsheck
2020 arXiv   pre-print
Finally, we conclude by proposing a novel auto-regressive model for capturing the intrinsic geometry and topology of data.  ...  , variational PDE modeling, and deep learning.  ...  Note that X α = (x α 1 − x α 0 , · · · , x α d − x α 0 ) ∈ R m×d is full rank matrix.  ... 
arXiv:2009.01786v1 fatcat:ohqjzwldqnhafmtsb57f7otfzu

Robust Subspace Recovery Layer for Unsupervised Anomaly Detection [article]

Chieh-Hsin Lai, Dongmian Zou, Gilad Lerman
2019 arXiv   pre-print
We propose a neural network for unsupervised anomaly detection with a novel robust subspace recovery layer (RSR layer).  ...  The encoder maps the data into a latent space, from which the RSR layer extracts the subspace.  ...  We thank Madeline Handschy for commenting on an earlier version of this paper.  ... 
arXiv:1904.00152v2 fatcat:bv6eso2mebgqhkbzr5iek4qbmi

OperatorNet: Recovering 3D Shapes From Difference Operators

Ruqi Huang, Marie-Julie Rakotosaona, Panos Achlioptas, Leonidas Guibas, Maks Ovsjanikov
2019 2019 IEEE/CVF International Conference on Computer Vision (ICCV)  
This paper proposes a learning-based framework for reconstructing 3D shapes from functional operators, compactly encoded as small-sized matrices.  ...  To this end we introduce a novel neural architecture, called OperatorNet, which takes as input a set of linear operators representing a shape and produces its 3D embedding.  ...  For example, several methods for shape interpolation have been proposed by designing deep neural networks, including auto-encoder architectures, and interpolating the latent vectors learned by such networks  ... 
doi:10.1109/iccv.2019.00868 dblp:conf/iccv/HuangRAGO19 fatcat:emllltoburehdgybsns4pudy3e

Deep Learning in Protein Structural Modeling and Design

Wenhao Gao, Sai Pooja Mahajan, Jeremias Sulam, Jeffrey J. Gray
2020 Patterns  
We dissect the emerging approaches using deep learning techniques for protein structural modeling and discuss advances and challenges that must be addressed.  ...  We argue for the central importance of structure, following the "sequence → structure → function" paradigm.  ...  However, MaSIF, similar to existing methods, showed low prediction accuracy for targets that involve conformational changes during binding.  ... 
doi:10.1016/j.patter.2020.100142 pmid:33336200 pmcid:PMC7733882 fatcat:qhjfidyusbfothv4kxdthodv6q

Face recognition via compact second order image gradient orientations [article]

He-Feng Yin, Xiao-Jun Wu, Xiaoning Song
2022 arXiv   pre-print
Experimental results indicate that the proposed method is superior to its competing approaches with few training samples, and even outperforms some prevailing deep neural network based approaches.  ...  Inspired by LRR and deep learning techniques, Xia et al. 22 developed an embedded conformal deep low-rank auto-encoder (ECLAE) neural network architecture for matrix recovery.  ...  To tackle the situation that both the training and test data are corrupted, low rank matrix recovery (LRMR) can be applied.  ... 
arXiv:2201.09246v2 fatcat:l3s2zqr3mrbwdkaz44awzqrhm4

TranAD: Deep Transformer Networks for Anomaly Detection in Multivariate Time Series Data [article]

Shreshth Tuli, Giuliano Casale, Nicholas R. Jennings
2022 arXiv   pre-print
In this paper, we propose TranAD, a deep transformer network based anomaly detection and diagnosis model which uses attention-based sequence encoders to swiftly perform inference with the knowledge of  ...  Despite the recent developments of deep learning approaches for anomaly detection, only a few of them can address all of these challenges.  ...  We thank Kate Highnam for constructive comments on improving the manuscript writing. We thank the providers of all datasets used in this work. We calculate the deviation as (𝑦 −𝑥)/𝑥.  ... 
arXiv:2201.07284v6 fatcat:eenjcq4u6bd6rk4vwxkts2gokm

Soft Autoencoder and Its Wavelet Adaptation Interpretation [article]

Fenglei Fan, Mengzhou Li, Yueyang Teng, Ge Wang
2021 arXiv   pre-print
Recently, deep learning becomes the main focus of machine learning research and has greatly impacted many important fields. However, deep learning is criticized for lack of interpretability.  ...  In this paper, we propose a new type of convolutional autoencoders, termed as Soft Autoencoder (Soft-AE), in which the activation functions of encoding layers are implemented with adaptable soft-thresholding  ...  The study in Ye et al. is most relevant to our work, in which the convolutional framelet theory with a low-rank Hankel matrix was leveraged to represent signals by their local and non-local bases, suggesting  ... 
arXiv:1812.11675v4 fatcat:6qc46dhblbhkbcqsmdf6gmywg4

LPD-AE: Latent Space Representation of Large-scale 3D Point Cloud

Chuanzhe Suo, Zhe Liu, Lingfei Mo, Yunhui Liu
2020 IEEE Access  
In this paper, we present a novel deep neural network, LPD-AE(Large-scale Place Description AutoEncoder Network), to obtain meaningful local and contextual features for the generation of latent space from  ...  The encoder network constructs the discriminative global descriptors to realize high accuracy and robust place recognition, which contributed by extracting the local neighbor geometric features and aggregating  ...  Although it exhibits the convincing feature extraction performance for place retrieval, it obtains low similarity of features extracted from similar places for the absence of decoder to recovery point  ... 
doi:10.1109/access.2020.2999727 fatcat:i55pbvejize3paprphcfwlki6a

UV-GAN: Adversarial Facial UV Map Completion for Pose-invariant Face Recognition [article]

Jiankang Deng, Shiyang Cheng, Niannan Xue, Yuxiang Zhou, Stefanos Zafeiriou
2017 arXiv   pre-print
In this paper, we propose a framework for training Deep Convolutional Neural Network (DCNN) to complete the facial UV map extracted from in-the-wild images.  ...  We demonstrate that by attaching the completed UV to the fitted mesh and generating instances of arbitrary poses, we can increase pose variations for training deep face recognition/verification models,  ...  Face representation using Deep Convolutional Neural Network (DCNN) embeddings is considered the method of choice for face verification, face clustering, and recognition [31, 28, 22, 30] .  ... 
arXiv:1712.04695v1 fatcat:bejoef247vfqpdc7afgli5hciu

UV-GAN: Adversarial Facial UV Map Completion for Pose-Invariant Face Recognition

Jiankang Deng, Shiyang Cheng, Niannan Xue, Yuxiang Zhou, Stefanos Zafeiriou
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
In this paper, we propose a framework for training Deep Convolutional Neural Network (DCNN) to complete the facial UV map extracted from in-the-wild images.  ...  We demonstrate that by attaching the completed UV to the fitted mesh and generating instances of arbitrary poses, we can increase pose variations for training deep face recognition/verification models,  ...  We thank the NVIDIA Corporation for the GPU donations.  ... 
doi:10.1109/cvpr.2018.00741 dblp:conf/cvpr/DengCXZZ18 fatcat:nuz3vfdgabdk5eggpefhzgt6ky
« Previous Showing results 1 — 15 out of 769 results