Sharing some of the latest research, announcements, and resources on deep learning.
Continuing our series of deep learning updates, we pulled together some of the awesome resources that have emerged since our last post. In case you missed it, here are our past updates: July, June (part 1, part 2, part 3, part 4), May, April (part 1, part 2), March part 1, February, November, September part 2 & October part 1, September part 1, August (part 1, part 2), July(part 1, part 2), June, and the original set of 20+ resources we outlined in April 2016. As always, this list is not comprehensive, so let us know if there’s something we should add, or if you’re interested in discussing this area further.
Research & Announcements
Cardiologist-Level Arrhythmia Detection With Convolutional Neural Networks by Rajpurkar et al of Stanford ML. We develop a model which can diagnose irregular heart rhythms, also known as arrhythmias, from single-lead ECG signals better than a cardiologist. Key to exceeding expert performance is a deep convolutional network which can map a sequence of ECG samples to a sequence of arrhythmia annotations along with a novel dataset two orders of magnitude larger than previous datasets of its kind. Original paper here.
Using Deep Learning to Create Professional-Level Photographs by Hui Wang of Google Research. Whether a photograph is beautiful or not is measured by its aesthetic value, which is a highly subjective concept. To explore how ML can learn subjective concepts, we introduce an experimental deep-learning system for artistic content creation. It mimics the workflow of a professional photographer, roaming landscape panoramas from Google Street View and searching for the best composition, then carrying out various postprocessing operations to create an aesthetically pleasing image. Original paper here.
How to turn audio clips into realistic lip-synced video by Suwajanakorn et al of University of Washington. Given audio of President Barack Obama, we synthesize a high quality video of him speaking with accurate lip sync, composited into a target video clip. Original paper here.
Introducing Vectorflow by Benoît Rostykus of Netflix. A lightweight neural network library for sparse data. A subset of our problems [at Netflix] involve dealing with extremely sparse data; the total dimensionality of the problem at hand can easily reach tens of millions of features, even though every observation may only have a handful of non-zero entries. For these cases, we felt the need for a minimalist library that is specifically optimized for training shallow feedforward neural nets on sparse data in a single-machine, multi-core environment.
Dota 2 bot by OpenAI. We’ve created a bot which beats the world’s top professionals at 1v1 matches of Dota 2 under standard tournament rules. The bot learned the game from scratch by self-play, and does not use imitation learning or tree search.
Finding Tiny Faces by Peiyun Hu and Deva Ramanan of Carnegie Mellon. We develop a face detector (Tiny Face Detector) that can find ~800 faces out of ~1000 reportedly present, by making use of novel characterization of scale, resolution, and context to find small objects. GitHub repo includes MATLAB implementation of Tiny face detector, including both training and testing code. A demo script is also provided. Original paper here.
Resources, Tutorials & Data
deeplearning.ai: Announcing new Deep Learning courses on Coursera by Andrew Ng. I have been working on three new AI projects, and am thrilled to announce the first one: deeplearning.ai, a project dedicated to disseminating AI knowledge, is launching a new sequence of Deep Learning courses on Coursera. These courses will help you master Deep Learning, apply it effectively, and build a career in AI.
Facets by Google. Visualization tools to better explore & understand machine learning data sets. The power of machine learning comes from its ability to learn patterns from large amounts of data. Understanding your data is critical to building a powerful machine learning system. Facets contains two robust visualizations to aid in understanding and analyzing machine learning datasets.
TensorFlow Neural Machine Translation Tutorial by Luong et al. This tutorial gives readers a full understanding of seq2seq models and shows how to build a competitive seq2seq model from scratch. We focus on the task of Neural Machine Translation (NMT) which was the very first testbed for seq2seq models with wild success.
The limitations of deep learning by Francois Chollet. The only real success of deep learning so far has been the ability to map space X to space Y using a continuous geometric transform, given large amounts of human-annotated data. Doing this well is a game-changer for essentially every industry, but it is still a very long way from human-level AI.
37 Reasons why your Neural Network is not working by Slav Ivanov. Insights and tips based on experience into why a network may not be training.
Cutting Edge Deep Learning For Coders, Part 2 by fast.ai. 7 week course on deep learning spanning artistic style, generative models, memory networks, attentional models, neural translation, and segmentation. Part 1, Practical Deep Learning for Coders, is here.
Heroes of Deep Learning by Andrew Ng. Andrew Ng interviews Geoffrey Hinton, of University of Toronto and Google. 40 min video.
An introduction to model ensembling by Jovan Sardinha. Model ensembling represents a family of techniques that help reduce generalization error in machine learning tasks. In this article, I will share some ways that ensembling has been employed and some basic intuition on why it works.
Neural networks for algorithmic trading: Multimodal and multitask deep learning by Alex Honchar. Tutorial on financial forecasting via artificial neural networks.
A 2017 Guide to Semantic Segmentation with Deep Learning by Sasank Chilamkurthy of Qure. Literature review and overview of semantic segmentation, i.e. understanding an image at a pixel level.