31st May - 13th June 2019
2747 new papers
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, CIFAR-10, ImageNet, COCO, and SVHN. Our trending phrases are Learned Optimization, Blue Noise, and Document Ranking. Survey and review papers are very popular in the past two weeks. In our trending papers section, we include an introductory paper on Variational Autoencoders, and a survey on Generative Adversarial Networks (GANs). We also include a paper on a new network design, Mesh R-CNN, to detect 2D images and produce the full 3D details. We also include two very interesting papers on robotics, to sum up the recent advancements in the robotics industry.Popular Datasets
Here are most mentioned datasets over the last two weeks.
| Name | Type | Number of Papers |
|---|---|---|
| MNIST | Handwritten Digits | 129 |
| CIFAR-10 | Tiny Image Dataset in 10 Classes | 76 |
| ImageNet | Image Dataset | 72 |
| COCO | Common Objects in Context | 31 |
| SVHN | The Street View House Numbers Dataset | 20 |
Trending Phrases
In this section, we present a list of phrases that appeared significantly more in this newsletter than the previous newsletters.
- Learned Optimization:
- Risks from Learned Optimization in Advanced Machine Learning Systems
- Understanding and correcting pathologies in the training of learned optimizers
- A Look at the Effect of Sample Design on Generalization through the Lens of Spectral Analysis
- A Passage-Based Approach to Learning to Rank Documents
Blue Noise:
Document Ranking:
Etymo Trending
Presented below is a list of the most trending papers added in the last two weeks.
-
An Introduction to Variational Autoencoders:
Variational autoencoders provide a principled framework for learning deep latent-variable models and corresponding inference models. In this work, we provide an introduction to variational autoencoders and some important extensions. -
Generative Adversarial Networks: A Survey and Taxonomy:
A study on the architecture-variants and loss-variants on Generative Adversarial Networks (GAN), and the proposal of using loss and architecture-variants for classifying most popular GANs. There is also a review of 16 GANs. -
Mesh R-CNN:
The authors unify advances in both 2D and 3D shape prediction. They propose a system that detects objects in real-world images and produces a triangle mesh giving the full 3D shape of each detected object. The system, called Mesh R-CNN, augments Mask R-CNN with a mesh prediction branch.
Two papers on Robotics
-
A Survey of Behavior Learning Applications in Robotics --
State of the Art and Perspectives:
A broad overview of behaviors that have been learned and used on real robots. The focus is on kinematically or sensorially complex robots, including humanoid robots or parts of humanoid robots, for example, legged robots or robotic arms. -
Air Learning: An AI Research Platform for Algorithm-Hardware
Benchmarking of Autonomous Aerial Robots:
An AI research platform for benchmarking algorithm-hardware performance and energy efficiency trade-offs. The focus is in particular on deep reinforcement learning (RL) interactions in autonomous unmanned aerial vehicles (UAVs). Users can specify a task, train different RL policies and evaluate their performance and energy efficiency on a variety of hardware platforms. More information and code for Air Learning can be found here.