Baidu built a supercomputer for deep learning

Gigaom

Chinese search engine company Baidu says it has built the world’s most-accurate computer vision system, dubbed Deep Image, which runs on a supercomputer optimized for deep learning algorithms. Baidu claims a 5.98 percent error rate on the ImageNet object classification benchmark; a team from Google won the 2014 ImageNet competition with a 6.66 percent error rate.

In experiments, humans achieved an estimated error rate of 5.1 percent on the ImageNet dataset.

The star of Deep Image is almost certainly the supercomputer, called Minwa, which Baidu built to house the system. Deep learning researchers have long (well, for the past few years) used GPUs in order to handle the computational intensity of training their models. In fact, the Deep Image research paper cites a study showing that 12 GPUs in a 3-machine cluster can rival the performance of the performance of the 1,000-node CPU cluster behind the famous Google Brain project, on which…

View original post 481 more words

Precision, Recall, AUCs and ROCs

The Shape of Data

I occasionally like to look at the ongoing Kaggle competitions to see what kind of data problems people are interested in (and the discussion boards are a good place to find out what techniques are popular.) Each competition includes a way of scoring the submissions, based on the type of problem. An interesting one that I’ve seen for a number of classification problems is the area under the Receiver Operating Characteristic (ROC) curve, sometimes shortened to the ROC score or AUC (Area Under the Curve). In this post, I want to discuss some interesting properties of this scoring system, and its relation to another similar measure – precision/recall.

View original post 2,354 more words

Hands on with Watson Analytics: Pretty useful when it’s working

Gigaom

Last month, [company]IBM[/company] made available the beta version of its Watson Analytics data analysis service, an offering first announced in September. It’s one of IBM’s only recent forays into anything resembling consumer software, and it’s supposed to make it easy for anyone to analyze data, relying on natural language processing (thus the Watson branding) to drive the query experience.

When the servers running Watson Analytics are working, it actually delivers on that goal.

Analytic power to the people

Because I was impressed that IBM decided to a cloud service using the freemium business model — and carrying the Watson branding, no less — I wanted to see firsthand how well Watson Analytics works. So I uploaded a CSV file including data from Crunchbase on all companies categorized as “big data,” and I got to work.

Seems like a good starting point.

watson14Choose one and get results. The little icon in…

View original post 433 more words

Facebook open sources tools for bigger, faster deep learning models

Gigaom

Facebook on Friday open sourced a handful of software libraries that it claims will help users build bigger, faster deep learning models than existing tools allow.

The libraries, which [company]Facebook[/company] is calling modules, are alternatives for the default ones in a popular machine learning development environment called Torch, and are optimized to run on [company]Nvidia[/company] graphics processing units. Among the modules are those designed to rapidly speed up training for large computer vision systems (nearly 24 times, in some cases), to train systems on potentially millions of different classes (e.g., predicting whether a word will appear across a large number of documents, or whether a picture was taken in any city anywhere), and an optimized method for building language models and word embeddings (e.g., knowing how different words are related to each other).

“‘[T]here is no way you can use anything existing” to achieve some of these results, said…

View original post 410 more words