Current methods for training convolutional neural networks depend on large amounts of labeled samples for supervised training. In this paper we present an approach for training a convolutional neural network using only unlabeled data. We train the network to discriminate between a set of surrogate classes. Each surrogate class is formed by applying a variety of transformations to a randomly sampled ’seed ’ image patch. We find that this simple feature learning algorithm is surprisingly successful when applied to visual object recognition. The feature representation learned by our algorithm achieves classification results matching or outperforming the current state-of-the-art for unsupervised learning on several popular datasets (STL-10, CIF...
Most recent unsupervised learning methods explore alternative objectives, often referred to as self-...
The goal of this work is to improve the robustness and generalization of deep learning models, using...
In this paper we study the problem of image representation learning without human annotation. By fol...
Current methods for training convolutional neural networks depend on large amounts of labeled sample...
Abstract—Deep convolutional networks have proven to be very successful in learning task specific fea...
Convolutional neural networks (CNNs) have achieved unprecedented success in a variety of computer vi...
Image classification is one of the core problems in Computer Vision. The classification task consist...
International audiencePre-training general-purpose visual features with convolutional neural network...
Abstract—Recognizing objects in natural images is an intricate problem involving multiple conflictin...
The latest generation of Deep Convolutional Neural Networks (DCNN) have dramatically advanced challe...
International audienceThis work proposes a new representation learning technique called convolutiona...
The concept of Convolution Neural Network (ConvNet or CNN) is evaluated from the animal visual corte...
Object recognition has been one of the main tasks in computer vision. While feature detection and cl...
Convolutional neural networks have shown remarkable ability to learn discriminative semantic feature...
Abstract In this paper, we propose a novel deep learning-based feature learning architecture for obj...
Most recent unsupervised learning methods explore alternative objectives, often referred to as self-...
The goal of this work is to improve the robustness and generalization of deep learning models, using...
In this paper we study the problem of image representation learning without human annotation. By fol...
Current methods for training convolutional neural networks depend on large amounts of labeled sample...
Abstract—Deep convolutional networks have proven to be very successful in learning task specific fea...
Convolutional neural networks (CNNs) have achieved unprecedented success in a variety of computer vi...
Image classification is one of the core problems in Computer Vision. The classification task consist...
International audiencePre-training general-purpose visual features with convolutional neural network...
Abstract—Recognizing objects in natural images is an intricate problem involving multiple conflictin...
The latest generation of Deep Convolutional Neural Networks (DCNN) have dramatically advanced challe...
International audienceThis work proposes a new representation learning technique called convolutiona...
The concept of Convolution Neural Network (ConvNet or CNN) is evaluated from the animal visual corte...
Object recognition has been one of the main tasks in computer vision. While feature detection and cl...
Convolutional neural networks have shown remarkable ability to learn discriminative semantic feature...
Abstract In this paper, we propose a novel deep learning-based feature learning architecture for obj...
Most recent unsupervised learning methods explore alternative objectives, often referred to as self-...
The goal of this work is to improve the robustness and generalization of deep learning models, using...
In this paper we study the problem of image representation learning without human annotation. By fol...