23rd April - 2nd May 2019
Etymo Newsletter provides the latest development in machine learning research,
including the most popular datasets and the most trending papers in the past two weeks.
If you like this newsletter, you can subscribe to our fortnightly newsletters here.
Fortnight Summary
Popular datasets include MNIST, ImageNet, COCO, and KITTI. Our trending phrases are Light Field, Truth Discovery, and Pairwise Learning. Three trending papers are related to GCNet, Unsupervised Data Augementation, and an image extractor using local retention networks. We also present you a collection of recent review papers in AI/Machine Learning.Popular Datasets
Here are most mentioned datasets over the last two weeks.
| Name | Type | Number of Papers |
|---|---|---|
| MNIST | Handwritten Digits | 58 |
| ImageNet | Image Dataset | 35 |
| CIFAR-10 | Tiny Image Dataset in 10 Classes | 30 |
| COCO | Common Objects in Context | 21 |
| KITTI | Autonomous Driving | 14 |
| Cityscapes | Images from 50 different cities | 11 |
| CIFAR-100 | Tiny Image Dataset in 100 Classes | 11 |
Trending Phrases
In this section, we present a list of phrases that appeared significantly more in this newsletter than the previous newsletters.
- Light Field:
- Local Light Field Fusion: Practical View Synthesis with Prescriptive Sampling Guidelines
- Using Social Network Information in Bayesian Truth Discovery
- Stability and Optimization Error of Stochastic Gradient Descent for Pairwise Learning
- Spectrum-enhanced Pairwise Learning to Rank
Truth Discovery:
Pairwise Learning:
Etymo Trending
Presented below is a list of the most trending papers added in the last two weeks.
-
GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond:
The authors designed a global context network (GCNet), which unifies both simplified NLNet and SENet and enhances it with a lightweight instantiation. GCNet generally outperforms both simplified NLNet and SENet on major benchmarks for various recognition tasks. The code and configurations are also available. -
Unsupervised Data Augmentation:
This paper proposes to apply data augmentation to unlabeled data in a semi-supervised learning setting. The method, named Unsupervised Data Augmentation or UDA, encourages the model predictions to be consistent between an unlabeled example and an augmented unlabeled example, by using state-of-the-art data augmentation methods. -
Local Relation Networks for Image Recognition:
The authors present a new image feature extractor, called the local relation layer, that adaptively determines aggregation weights based on the compositional relationship of local pixel pairs. It can composite visual elements into higher-level entities in a more efficient manner that benefits semantic inference.