NeurIPS-2017 Paper Digest

Every paper, ordered by impact or potential impact. About NeurIPS 2017

IDPaperAuthors
1Attention Is All You Need
IF:10   Related Papers   Related Patents   Related Grants   Related Venues   Related Experts   Related Code   View    Save
Highlight: We propose a novel, simple network architecture based solely onan attention mechanism, dispensing with recurrence and convolutions entirely.Experiments on two machine translation tasks show these models to be superiorin quality while being more parallelizable and requiring significantly less timeto train.
Ashish Vaswani; Noam Shazeer; Niki Parmar et al.
2GANs Trained By A Two Time-Scale Update Rule Converge to A Local Nash Equilibrium
IF:10   Related Papers   Related Patents   Related Grants   Related Venues   Related Experts   Related Code   View    Save
Highlight: We propose a two time-scale update rule (TTUR) for training GANs with stochastic gradient descent on arbitrary GAN loss functions.
Martin Heusel; Hubert Ramsauer; Thomas Unterthiner et al.
3Improved Training of Wasserstein GANs
IF:10   Related Papers   Related Patents   Related Grants   Related Venues   Related Experts   Related Code   View    Save
Highlight: We propose an alternative to clipping weights: penalize the norm of gradient of the critic with respect to its input.
Ishaan Gulrajani; Faruk Ahmed; Martin Arjovsky et al.
4A Unified Approach to Interpreting Model Predictions
IF:10   Related Papers   Related Patents   Related Grants   Related Venues   Related Experts   Related Code   View    Save
Highlight: To address this problem, we present a unified framework for interpreting predictions, SHAP (SHapley Additive exPlanations).
Scott M. Lundberg; Su-In Lee;
5PointNet++: Deep Hierarchical Feature Learning on Point Sets in A Metric Space
IF:10   Related Papers   Related Patents   Related Grants   Related Venues   Related Experts   Related Code   View    Save
Highlight: In this work, we introduce a hierarchical neural network that applies PointNet recursively on a nested partitioning of the input point set. With further observation that point sets are usually sampled with varying densities, which results in greatly decreased performance for networks trained on uniform densities, we propose novel set learning layers to adaptively combine features from multiple scales.
Charles Ruizhongtai Qi; Li Yi; Hao Su et al.

About NeurIPS 2017