Neural Machine Translation (NMT) has achieved remarkable progress with the quick evolvement of model structures. In this paper, we propose the concept of layer-wise coordination for NMT, which explicitly coordinates the learning of hidden representations of the encoder and decoder together layer by layer, gradually from low level to high level.

We argue that this GAN-based approach can help few-shot classifiers to learn sharper decision boundary, which could generalize better.

Stochastic Gradient Descent (SGD) and its variants are the current workhorse for training neural networks.

When workers cannot form a fully connected communication topology or the communication latency is high (e.g., in

Neural Information Processing Systems http://nips.cc/

We propose a structure-adaptive variant of a state-of-the-art stochastic variance-reduced gradient algorithm Katyusha for regularized empirical risk minimization.

Clustering, and in particular center based clustering, are central problems in unsupervised learning.

Scaling model capacity has been vital in the success of deep learning.

We present a new algorithm to generate minimal, stable, and symbolic corrections to an input that will cause a neural network with ReLU activations to change its output.

Surprisingly we show that volume sampling can have poor behavior when we require a very accurate approximation - indeed worse than some i.i.d.