GradientBased Learning Applied to Document Recognition

By Simon S. Haykin, Bart Kosko. 2001

4,386 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Optimization & Training

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Optimization & Training

Citations in the corpus

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation (2013, background): “Regardless of the size or aspect ratio of the candidate region, we warp all pixels in a tight bounding box around it to the required size.”
Measuring the Carbon Intensity of AI in Cloud Instances (2022, methodology): “We trained three sizes of DenseNets [19] onMNIST [25].”
Attention Augmented Convolutional Networks (2019, background): “Convolutional Neural Networks have enjoyed tremendous success in many computer vision applications, especially in image classification [24, 23].”
Making Convolutional Networks Shift-Invariant Again (2019, background): “Local connectivity and weight sharing have been a central tenet of neural networks, including the Neocognitron (Fukushima & Miyake, 1982), LeNet (LeCun et al., 1998) and modern networks such as Alexnet (Krizhevsky et al.”
Zero-Shot Text-to-Image Generation (2021, methodology): “Architecture The dVAE encoder and decoder are convolutional (LeCun et al., 1998) ResNets (He et al., 2016) with bottleneck-style resblocks.”
Stochastic Normalizing Flows (2020, background): “MNIST and Fashion-MNIST VAE in Table 1 • The latent space dimension was set to 50.”
BIVA: A Very Deep Hierarchy of Latent Variables for Generative Modeling (2019, methodology): “We evaluate BIVA L = 6 in terms of test log-likelihood on statically binarized MNIST (Salakhutdinov & Murray, 2008), dynamically binarized MNIST (LeCun et al., 1998) and dynamically binarized OMNIGLOT (Lake et al.”
Learning Generative ConvNets via Multi-grid Modeling and Sampling (2017, methodology): “It can be derived from the commonly used discriminative ConvNet [23, 21] as a direct consequence of the Bayes rule [4], but unlike the discriminative ConvNet, the generative ConvNet is endowed with the gift of imagination in that it can generate images by sampling from the probability distribution o”
Perceiver: General Perception with Iterative Attention (2021, background): “ConvNets (Fukushima, 1980; LeCun et al., 1998; Cires¸an et al., 2011; Krizhevsky et al., 2012) have been the dominant family of architectures for perceptual tasks for nearly a full decade, thanks to their good performance and scalability.”
Pedestrian Detection with Unsupervised Multi-stage Feature Learning (2012, background): “A randomly selected 20% of the connections in mapping from input features to output features is removed to limit the computational requirements and break the symmetry [23].”
From Pixels to Torques: Policy Learning with Deep Dynamical Models (2015, background): “…architectures, such as deep neural networks (Hinton & Salakhutdinov, 2006), stacked auto-encoders (Bengio et al., 2007; Vincent et al., 2008), or convolutional neural networks (LeCun et al., 1998), are the current state of the art in learning parsimonious representations of high-dimensional data.”
Geometric Deep Learning on Graphs and Manifolds Using Mixture Model CNNs (2016, methodology): “Three methods were compared: classical CNN LeNet5 architecture [23] (containing two convolutional, two maxpooling, and one fully-connected layer, applied on regular grids only), spectral ChebNet[13] and the proposed MoNet.”
Improved Regularization of Convolutional Neural Networks with Cutout (2017, methodology): “Much of these improvements can be attributed to the use of convolutional neural networks (CNNs) [9], which are capable of learning complex hierarchical feature representations of images.”
Generative Moment Matching Networks (2015, methodology): “On MNIST and the Toronto Face Dataset (TFD) we demonstrate improved results over comparable baselines, including GANs.”
Unsupervised Domain Adaptation by Backpropagation (2014, methodology): “Below, we detail the proposed approach to domain adaptation in deep architectures, and present results on traditional deep learning image datasets (such as MNIST (LeCun et al., 1998) and SVHN (Netzer et al.”
The Power of Interpolation: Understanding the Effectiveness of SGD in Modern Over-parametrized Learning (2017, neutral): “This is consistent with our experimental results for MNIST (LeCun et al., 1998), HINTS (Healy et al.”
Value Iteration Networks (2016, background): “We now have all the ingredients for a differentiable planning-based policy, which we term a value iteration network (VIN).”
Deep Visual-Semantic Alignments for Generating Image Descriptions (2014, neutral): “We then treat these correspondences as training data for our multimodal Recurrent Neural Network model which learns to generate the descriptions.”