Machine Learning Roundup: 10/6/2016
Machine learning (ML) and deep learning (DL) content from the past 24 hours, plus a few older items.
Gale Morrison’s survey of neural computing in Semiconductor Engineering is an absolute must read. She summarizes intellectual and technical developments in the field, market drivers and briefly discusses what works. Companies mentioned: Amazon, Baidu, Google, Huawei, Intel, Nervana, NVIDIA, and Samsung.
In Forbes, Karl Freund describes MSFT’s interest in FPGA accelerators.
Dr. Diana Maynard writes how to teach sarcasm to bots. Exactly what we need.
In the Fiscal Times, Frank Konkel summarizes how the CIA uses machine learning to predict such things as the flow of illicit cash to extremists. In Defense One, Frank Konkel reports that the CIA says it can predict social unrest 3-5 days in advance. I don’t think this is what folks have in mind when they call for Data Science for Social Good.
NVIDIA continues to eat the headlines. AI and robotics dominate NVIDIA’s GTC Japan show this week. UK-based computer retailer Scan proposes to offer time on NVIDIA’s stupidly powerful DGX-1, thereby demonstrating that GPU-based systems are the new mainframes. 🙂 Wall Street notices that NVIDIA is on a roll; in Seeking Alpha, Chris Lau explains. Meanwhile, HPC cloud provider Nimbix announces a partnership with NVIDIA and IBM; Nimbix will power its cloud with IBM Power Systems S822LC servers featuring NVIDIA Pascal GPUs. According to reports, AMD’s RTG Radeon Technology will market a dual-GPU chip in 2017. Linkapalooza here.
Killing Data Scientists
On a related topic, Richard Boire wonders if deep learning will kill feature engineering, a fancy term for “manually mucking around with the data to get learning algorithms to work.” In theory, deep learning algorithms do that mucking around for you. The key word there is “theory.”
ML in Healthcare
Jennifer Bresnick summarizes the growing role of ML and AI in Healthcare
New Zealand-based Orion Health publishes a report summarizing the role of machine learning in healthcare. Orion was just named as a partner in a large-scale field test of precision medicine.
AI/DL/ML for Executives
Royal Frasier opines on the future of AI.
In Fortune, Geoff Colvin explains why executives must understand AI/ML/DL. In typical fashion for writers at this level, he muddles AI, DL, and ML together into a blob. Hint for executives: don’t do that.
Scott Stiner, writing in Forbes, explains how to set up an ML framework for your company.
Alpine’s Steven Hillion and Pivotal’s Kaushik Das collaborate on a piece warning executives to focus on business outcomes and not the details of Big Data infrastructure. They cite some interesting statistics. According to McKinsey, 86% of executives deem their analytics programs only somewhat effective. A Forrester report notes that while 74% of enterprise architects aspire to be data-driven, just 29% say their firms are good at translating analytics into business results.
Will Rinehart of the American Action Forum — whatever that is — opines that “algorithms” are important to public policy. That means he wants the government to regulate “algorithms.” He admits that the word “algorithm” is often confused with ML and AI, then adds to the confusion.
Microsoft Pushes Something
There is a slew of stories today on MSFT, which means a PR operation. I don’t see much organic news, but here’s a linkapalooza.
Sofia Tomov, age 12, develops an algorithm to stop adverse reactions to prescription meds.
Udacity releases more driving data to open source.
Ozzie football team uses AI to analyze play.
Nugat, a Singapore-based provider of natural language and vision processing technology, raises $5.2 million.