All Research Areas
Research Areas
Year Published

127 Results

July 21, 2017

Semantic Amodal Segmentation

CVPR 2017

Common visual recognition tasks such as classification, object detection, and semantic segmentation are rapidly reaching maturity, and given the recent rate of progress, it is not unreasonable to conjecture that techniques for many of these problems will approach human levels of performance in the next few years. In this paper we look to the future: what is the next frontier in visual recognition?

By: Yan Zhu, Yuandong Tian, Dimitris Mexatas, Piotr Dollar
July 21, 2017

Aggregated Residual Transformations for Deep Neural Networks

CVPR 2017

We present a simple, highly modularized network architecture for image classification.

By: Saining Xie, Ross Girshick, Piotr Dollar, Zhuowen Tu, Kaiming He
July 21, 2017

Densely Connected Convolutional Networks

CVPR 2017

In this paper, we embrace the observation that hat convolutional networks can be substantially deeper, more accurate, and efficient to train if they contain shorter connections between layers close to the input and those close to the output, and introduce the Dense Convolutional Network (DenseNet), which connects each layer to every other layer in a feed-forward fashion.

By: Gao Huang, Zhuang Liu, Laurens van der Maaten, Kilian Q. Weinberger
June 8, 2017

Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour

Data @ Scale

In this paper, we empirically show that on the ImageNet dataset large minibatches cause optimization difficulties, but when these are addressed the trained networks exhibit good generalization.

By: Priya Goyal, Piotr Dollar, Ross Girshick, Pieter Noordhuis, Lukasz Wesolowski, Aapo Kyrola, Andrew Tulloch, Yangqing Jia, Kaiming He
May 21, 2017

CAN: Creative Adversarial Networks

IEEE International Conference on Communications (ICCC)

We propose a new system for generating art. The system generates art by looking at art and learning about style; and becomes creative by increasing the arousal potential of the generated art by deviating from the learned styles. We build over Generative Adversarial Networks (GAN), which have shown the ability to learn to generate novel images simulating a given distribution.

By: Ahmed Elgammal, Bingchen Liu, Mohamed Elhoseiny, Marian Mazzone
April 24, 2017

Unsupervised Cross-Domain Image Generation

International Conference on Learning Representations (ICLR) 2017

We study the problem of transferring a sample in one domain to an analog sample in another domain. Given two related domains, S and T, we would like to learn a generative function G that maps an input sample from S to the domain T, such that the output of a given representation function f, which accepts inputs in either domains, would remain unchanged.

By: Yaniv Taigman, Adam Polyak, Lior Wolf
April 24, 2017

An Analytical Formula of Population Gradient for Two-Layered ReLU network and its Applications in Convergence and Critical Point Analysis

International Conference on Learning Representations (ICLR) 2017

In this paper, we explore theoretical properties of training a two-layered ReLU network g(x; w) = PK j=1 σ(w | j x) with centered d-dimensional spherical Gaussian input x (σ=ReLU). We train our network with gradient descent on w to mimic the output of a teacher network with the same architecture and fixed parameters w∗.

By: Yuandong Tian
April 24, 2017

LR-GAN: Layered Recursive Generative Adversarial Networks for Image Generation

International Conference on Learning Representations (ICLR)

We present LR-GAN: an adversarial image generation model which takes scene structure and context into account.

By: Jianwei Yang, Anitha Kannan, Dhruv Batra, Devi Parikh
April 24, 2017

Towards Principled Methods for Training Generative Adversarial Networks

International Conference on Learning Representations (ICLR) 2017

The goal of this paper is not to introduce a single algorithm or method, but to make theoretical steps towards fully understanding the training dynamics of generative adversarial networks.

By: Leon Bottou, Martin Arjovsky
April 24, 2017

Improving Neural Language Models with a Continuous Cache

International Conference on Learning Representations (ICLR) 2017

We propose an extension to neural network language models to adapt their prediction to the recent history. Our model is a simplified version of memory augmented networks, which stores past hidden activations as memory and accesses them through a dot product with the current hidden activation.

By: Armand Joulin, Edouard Grave, Nicolas Usunier