Roundup 11/15/2016

Machine learning (ML) and deep learning (DL) content from the past 24 hours. Plus, a few leftovers from Friday and the weekend.

In Bloomberg, Nishant Kumar explains why machines “still can’t learn so good.” The jokes write themselves.

According to the Official Microsoft Blog, Microsoft researchers recently discovered lung cancer risks in web search logs. So, you should definitely avoid them.

Fundamentals

— In The Register, Dr. Stephen Perrenod delivers a concise overview of AI, ML, and DL, covering everything from applications to hardware in 600 words.

— In InfoWorld, David Linthicum tries to explain how to approach machine learning in the cloud. Unfortunately, he confuses machine inference with machine learning.

— Serge Haziyev of SoftServe describes the progress of machine learning over the past thirty years.

Research

— A team at Ohio State University announces the High Performance Deep Learning project, which aims to accelerate DL with HPC.

Methods and Techniques

— Matthew Mayo explains ensemble learners.

Competitions

— On Kaggle, The Nature Conservatory launches a competition to detect and classify fish. In Fortune, Jonathan Vanian provides background. First prize is $50K. Entry deadline April 5, 2017.

Software/Services

— On the AWS Compute Blog, Naveen Swamy and Joseph Spisak explain distributed deep learning with the AWS Deep Learning AMI. The AMI supports Caffe, CNTK/Microsoft Cognitive, MXNet, TensorFlow, Theano, and Torch. Product reviewers are underwhelmed, noting the absence of NVIDIA, CUDA, Anaconda or Python3 support.

— On Slideshare, Makoto Yui describes Apache Hivemall, a machine learning library for Hive, Pig, and Spark.

— H2O.ai posts videos from the recent H2O Open Tour Dallas event.

Hardware

— Timothy Prickett Morgan examines NVIDIA’s recent growth.

— Also on The Next Platform, Nicole Hemsoth dissects Cray’s new XC50 supercomputer, built on the NVIDIA Pascal GPU-accelerated chip.

— IBM launches the PowerAI software kit to run on OpenPOWER LC servers for DL/AI workloads.

Applications

— In Futurism, Jelor Gallego describes TeraStructure, an algorithm that analyzes genome data at scale.

— Scientists use ML to detect feeding patterns for insects that carry disease.

— Digital media agency Zenith uses ML to assist media planners.

— Microsoft introduces a platform (branded as Decision Service) that uses ML to optimize content, pricing, and ad serving in real time.

— Uber uses ML to make its app smarter.

— The Allen Institute for Artificial Intelligence introduces a search engine (Semantic Scholar) that uses ML to improve searches through academic research.

— Zillow develops a deep neural network to detect specific attributes in a home from digital images.

— Google Play uses machine learning to predict what you want to hear. Like Spotify, but not as good.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s