Modeling and generating graphs is fundamental for studying networks in biology, engineering, and social sciences. However, modeling complex distributions over graphs and then efficiently sampling from these distributions is challenging due to the non-unique, high-dimensional nature of graphs and the complex, non-local dependencies that exist between edges in a given graph. Here we propose GraphRNN, a deep autoregressive model that addresses the above challenges and approximates any distribution of graphs with minimal assumptions about their structure. GraphRNN learns to generate graphs by training on a representative set of graphs and decomposes the graph generation process into a sequence of node and edge formations, conditioned on the graph structure generated so far. In order to quantitatively evaluate the performance of GraphRNN, we introduce a benchmark suite of datasets, baselines and novel evaluation metrics based on Maximum Mean Discrepancy, which measure distances between sets of graphs. Our experiments show that GraphRNN significantly outperforms all baselines, learning to generate diverse graphs that match the structural characteristics of a target set, while also scaling to graphs 50 times larger than previous deep models.

Many problems in machine learning and related application areas are fundamentally variants of conditional modeling and sampling across multi-aspect data, either multi-view, multi-modal, or simply multi-group. For example, sampling from the distribution of English sentences conditioned on a given French sentence or sampling audio waveforms conditioned on a given piece of text. Central to many of these problems is the issue of missing data: we can observe many English, French, or German sentences individually but only occasionally do we have data for a sentence pair. Motivated by these applications and inspired by recent progress in variational autoencoders for grouped data [1], we develop factVAE, a deep generative model capable of handling multi-aspect data, robust to missing observations, and with a prior that encourages disentanglement between the groups and the latent dimensions. The effectiveness of factVAE is demonstrated on a variety of rich real-world datasets, including motion capture poses and pictures of faces captured from varying poses and perspectives.

There are so many tools, platforms and resources available, MLEs can focus their time on solving problems critical to their field or company instead of worrying about building platforms and hand rolling numerical algorithms. Google Cloud has easy means of building and deploying TensorFlow models including their new TPU support in beta, AWS has an ever evolving suite of deep learning AMIs and Nvidia has a great deep learning SDK. In parallel, Apple's coreML and Android's NN API make is simpler and faster to deploy models on phones; this will continue to push the boundary for developing and releasing ML apps. With all of the above, there is healthy competition among big players in the cloud space pushing the whole ecosystem forward. And yet, most of them are finding ways to collaborate towards open standards like ONNX.

A quiet wager has taken hold among researchers who study artificial intelligence techniques and the societal impacts of such technologies. They're betting whether or not someone will create a so-called Deepfake video about a political candidate that receives more than 2 million views before getting debunked by the end of 2018. The actual stakes in the bet are fairly small: Manhattan cocktails as a reward for the "yes" camp and tropical tiki drinks for the "no" camp. But the implications of the technology behind the bet's premise could potentially reshape governments and undermine societal trust in the idea of having shared facts. It all comes down to when the technology may mature enough to digitally create fake but believable videos of politicians and celebrities saying or doing things that never actually happened in real life.

When it comes to groundbreaking research, there are two fields that seem to occupy the newscycle: carbon nanotubes and artificial intelligence. The potential combination of those two fields of study seems like it could radically change the word as we know it, or, as South Korean scientists have discovered, at least change how we type. The carbon atom, one of the building blocks of life, gains radical new abilities when assembled into long, thin chains, known as carbon nanotubes. Think ultra-flexible films that are better at stopping bullets than kevlar vests, or bio-engineered plants that can detect land mines and explosives. And AI, trained using deep learning techniques, is soon going to make it almost impossible to discern fake videos from real ones. But researchers from South Korea's Sejong University, Chung-Ang University, and Kyungpook National University are instead merging those burgeoning technologies to create an ultra-thin portable keyboard that can be crumpled up like paper without breaking it.

Toronto is a thriving hub for AI experts, thanks in part to foundational work out of the University of Toronto and government-supported research organizations like the Vector Institute. We're tapping further into this expertise by investing in a new AI research lab -- led by leading computer scientist Sanja Fidler -- that will become the focal point of our presence in the city. NVIDIA's Toronto office opened in 2015, leveraging our acquisition of TransGaming, a game-technology company, and currently numbers about 50. With the new lab, our goal is to triple the number of AI and deep learning researchers working there by year's end. It will be a state-of-art facility for AI talent to work in and will expand the footprint of our office by about half to accommodate the influx of talent.

Anyone who has lived through the 1980s knows how maddeningly difficult it is to solve a Rubik's Cube, and to accomplish the feat without peeling the stickers off and rearranging them. Apparently the six-sided contraption presents a special kind of challenge to modern deep learning techniques that makes it more difficult than, say, learning to play chess or Go. That used to be the case, anyway. Researchers from the University of California, Irvine, have developed a new deep learning technique that can teach itself to solve the Rubik's Cube. What they come up with is very different than an algorithm designed to solve the toy from any position. It's easy enough to take such an algorithm and then automate the process, but that kind of solution is limited in scope and utility.

This is the second article on my series introducing machine learning concepts with while stepping very lightly on mathematics. If you missed previous article you can find in here (on KL divergence). Fun fact, I'm going to make this an interesting adventure by introducing some machine learning concept for every letter in the alphabet (This would be for the letter C). Convolution neural networks (CNNs) are a family of deep networks that can exploit the spatial structure of data (e.g. Think of a problem where we want to identify if there is a person in a given image.

Skeletal bone age assessment is a common clinical practice to diagnose endocrine and metabolic disorders in child development. In this paper, we describe a fully automated deep learning approach to the problem of bone age assessment using data from the 2017 Pediatric Bone Age Challenge organized by the Radiological Society of North America. The dataset for this competition is consisted of 12.6k radiological images. Each radiograph in this dataset is an image of a left hand labeled by the bone age and the sex of a patient. Our approach utilizes several deep neural network architectures trained end-to-end.

Deep learning has changed the way we work, compute and has made our lives a lot easier. As Andrej Karpathy mentioned it is indeed the software 2.0, as we have taught machines to figure things out themselves. There are many existing deep learning techniques which can be ascribed to its prolific success. But no major impact has been created by deep generative models, which is due to their inability to approximate intractable probabilistic computations. Ian Goodfellow was able to find a solution that could sidestep these difficulties faced by generative models and created a new ingenious model called Generative Adversarial Networks. I believe it is astonishing when you look at the capabilities of a GAN.