• DnCNN

  • Referenced in 45 articles [sw39678]
  • progress in very deep architecture, learning algorithm, and regularization method into image denoising. Specifically, residual...
  • PointNet

  • Referenced in 43 articles [sw31209]
  • PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation. Point cloud ... format, most researchers transform such data to regular 3D voxel grids or collections of images...
  • Xception

  • Referenced in 23 articles [sw39068]
  • Xception: Deep Learning with Depthwise Separable Convolutions. We present an interpretation of Inception modules ... being an intermediate step in-between regular convolution and the depthwise separable convolution operation ... observation leads us to propose a novel deep convolutional neural network architecture inspired by Inception...
  • NETT

  • Referenced in 15 articles [sw41773]
  • there are few theoretical results for deep learning in inverse problems. In this paper ... consistent solutions having small value of a regularizer defined by a trained neural network...
  • Drop-Activation

  • Referenced in 1 article [sw41511]
  • parameter reduction and harmonious regularization. Overfitting frequently occurs in deep learning. In this paper ... activations randomly. Our theoretical analyses support the regularization effect of drop-activation as implicit parameter ... extit{C. Szegedy}, “Batch normalization: accelerating deep network training by reducing internal covariate shift”, Preprint ... regularizer drop-activation can be used in harmony with standard training and regularization techniques such...
  • BinaryConnect

  • Referenced in 24 articles [sw35871]
  • development of dedicated hardware for Deep Learning (DL). Binary weights, i.e., weights which are constrained ... schemes, we show that BinaryConnect acts as regularizer and we obtain near state...
  • TorchDyn

  • Referenced in 1 article [sw35071]
  • emerged as a novel perspective on deep learning, improving performance in tasks related to dynamical ... accessible as regular plug-and-play deep learning primitives. This objective is achieved by identifying...
  • mixup

  • Referenced in 11 articles [sw35857]
  • mixup: Beyond Empirical Risk Minimization. Large deep neural networks are powerful, but exhibit undesirable behaviors ... this work, we propose mixup, a simple learning principle to alleviate these issues. In essence ... their labels. By doing so, mixup regularizes the neural network to favor simple linear behavior...
  • RGCNN

  • Referenced in 1 article [sw36662]
  • RGCNN: Regularized Graph CNN for Point Cloud Segmentation. Point cloud, an efficient 3D object representation ... data format, previous deep learning works often convert point clouds to regular 3D voxel grids...
  • ISTA-Net

  • Referenced in 7 articles [sw40591]
  • reconstruction model. To cast ISTA into deep network form, we develop an effective strategy ... proximal mapping associated with the sparsity-inducing regularizer using nonlinear transforms. All the parameters ... transforms, shrinkage thresholds, step sizes, etc.) are learned end-to-end, rather than being hand...
  • VoxSegNet

  • Referenced in 1 article [sw36661]
  • deep learning in shape analysis due to its generalization ability and regular data format. However...
  • Albumentations

  • Referenced in 2 articles [sw32338]
  • become a common implicit regularization technique to combat overfitting in deep convolutional neural networks ... used to improve performance. While most deep learning frameworks implement basic image transformations, the list...
  • Quicksilver

  • Referenced in 7 articles [sw38623]
  • Quicksilver: Fast predictive image registration – A deep learning approach. This paper introduces Quicksilver, a fast ... model based directly on image appearance. A deep encoder-decoder network is used ... guaranteed diffeomorphic mappings for sufficiently strong regularization. We also provide a probabilistic version ... four standard validation datasets, and can jointly learn an image similarity measure. Quicksilver is freely...
  • Deepr

  • Referenced in 1 article [sw40726]
  • predictive regular clinical motifs from irregular episodic records. We present Deepr (short for Deep ... record), a new end-to-end deep learning system that learns to extract features from...
  • SoftRegex

  • Referenced in 2 articles [sw36202]
  • reinforced learning based on the semantic (rather than syntactic) equivalence between two regular expressions. Since ... regular expression equivalence problem is PSPACE-complete, we introduce the EQ_Reg model for computing ... simi-larity of two regular expressions using deep neural networks...
  • TFLearn

  • Referenced in 1 article [sw21054]
  • TFlearn is a modular and transparent deep learning library built on top of Tensorflow ... understand high-level API for implementing deep neural networks, with tutorial and examples. Fast prototyping ... highly modular built-in neural network layers, regularizers, optimizers, metrics... Full transparency over Tensorflow ... currently supports most of recent deep learning models, such as Convolutions, LSTM, BiRNN, BatchNorm, PReLU...
  • PointConv

  • Referenced in 2 articles [sw41516]
  • Clouds. Unlike images which are represented in regular dense grids, 3D point clouds are irregular ... applied on point clouds to build deep convolutional networks. We treat convolution kernels as nonlinear ... given point, the weight functions are learned with multi-layer perceptron networks and density functions...
  • RandAugment

  • Referenced in 2 articles [sw40532]
  • significantly improve the generalization of deep learning models. Recently, automated augmentation strategies have ... results in semi-supervised learning and improved robustness to common corruptions of images. An obstacle ... these approaches are unable to adjust the regularization strength based on model or dataset size...
  • CAZSL

  • Referenced in 1 article [sw38827]
  • study the problem of designing deep learning agents which can generalize their models ... physical world by building context-aware learning models. The purpose of these agents ... motivation, we present context-aware zero shot learning (CAZSL, pronounced as casual) models, an approach ... Siamese network architecture, embedding space masking and regularization based on context variables which allows...
  • ANSYS

  • Referenced in 704 articles [sw00044]
  • ANSYS offers a comprehensive software suite that spans...