Why you can’t calculate aging trajectories with a standard regression

I found myself in a little Twitter discussion last week about using regression to analyze player aging. I argued that regression won’t give you accurate results, and that the less elegant “delta method” is the better way to go.Although I did a small example to try to make my point, Tango suggested I do a bigger simulation and a blog post. That’s this.

Read Full Story

Applying NLP and Entity Extraction To The Russian Twitter Troll Tweets In Neo4j (and more Python!)

Russian Twitter Troll Tweets
Previously, we explored how to scrape tweets from Internet Archive that have been removed from Twitter.com and the Twitter API as a result of the US House Intelligence Committee’s investigation into Russia’s involvement in influencing the 2016 election through social media, largely by spreading fake news.

Read Full Story

Scraping Russian Twitter Trolls With Python, Neo4j, and GraphQL

Last week as a result of the House Intelligence Select Committee investigation, Twitter released the screen names of 2752 Twitter accounts tied to Russia’s Internet Research Agency that were involved in spreading fake news, presumably with the goal of influencing the 2016 election.

Read Full Story

Applying NLP and Entity Extraction To The Russian Twitter Troll Tweets In Neo4j (and more Python!) · William Lyon

Russian Twitter Troll Tweets
Previously, we explored how to scrape tweets from Internet Archive that have been removed from Twitter.com and the Twitter API as a result of the US House Intelligence Committee’s investigation into Russia’s involvement in influencing the 2016 election through social media, largely by spreading fake news.

Read Full Story

Scraping Russian Twitter Trolls With Python, Neo4j, and GraphQL · William Lyon

Last week as a result of the House Intelligence Select Committee investigation, Twitter released the screen names of 2752 Twitter accounts tied to Russia’s Internet Research Agency that were involved in spreading fake news, presumably with the goal of influencing the 2016 election.

Read Full Story

Building a business around Machine Learning APIs

I got a variety of reactions on Twitter following my GigaOM piece on how Data Scientists work at automating themselves. One of them I want to discuss today is about building businesses on top / around Prediction APIs such as Google’s or BigML’s (a.k.a. machine learning APIs).

Read Full Story

Filtering for English Tweets: Unsupervised Language Detection on Twitter

(See a demo here.)
While working on a Twitter sentiment analysis project, I ran into the problem of needing to filter out all non-English tweets. (Asking the Twitter API for English-only tweets doesn’t seem to work, as it nonetheless returns tweets in Spanish, Portuguese, Dutch, Russian, and a couple other languages.

Read Full Story

Fast.ai courses: Part 2: Deep Learning from the Foundations

** Nuit Blanche is now on Twitter: @NuitBlog ** One of the very appealing aspects of Machine learning these days is the ability to learn from very good people. Jeremy Howard at fast.ai does that very well and he just announced a new course entitled Part 2: Deep Learning from the Foundations.

Read Full Story

Improving Neural Architecture Search Image Classifiers via Ensemble Learning – implementation –

** Nuit Blanche is now on Twitter: @NuitBlog ** Improving Neural Architecture Search Image Classifiers via Ensemble Learning by Vladimir Macko, Charles Weill, Hanna Mazzawi and Javier Gonzalvo‎AdaNAS is an algorithm for learning an ensemble that improves the performance of neural architecture search models while having a similar parameter count as single large model.

Read Full Story

Degrees of Freedom Analysis of Unrolled Neural Networks

** Nuit Blanche is now on Twitter: @NuitBlog ** Studying the great convergence ! Degrees of Freedom Analysis of Unrolled Neural Networks by Morteza Mardani, Qingyun Sun, Vardan Papyan, Shreyas Vasanawala, John Pauly, David DonohoUnrolled neural networks emerged recently as an effective model for learning inverse maps appearing in image restoration tasks. However, their generalization risk (i.e.

Read Full Story

the psychology of the enterprise buyer

Consumer startups like Facebook, Twitter, Pinterest, and even DropBox are built by founders who wanted to “make something cool” for their own benefit. Their teams intuitively understand what works because they are their own target audience: young, tech-savvy people looking for better ways to connect, share, and organize their digital stuff.

Read Full Story

ICML 2012 Deep Learning and Unsupervised Feature Extraction Reading List

The ICML2012 accepted papers are officially online.On twitter, Andrej Kaparthy complained that the list is a bit hard to browse through. I agree and even though this is probably not the nice visualization he had in mind, I felt like having topical reading lists would somehow mitigate this problem.

Read Full Story

Courses: Marc Lelarge’s hands-on tour to deep learning with PyTorch

** Nuit Blanche is now on Twitter: @NuitBlog ** From Marc’s twitter account:Just finished teaching a 5 days summer school: Hands-on tour to deep learning with @PyTorch (https://t.co/LjFjnTGa5j), in collaboration with @abursuc, @tlacroix6 and Stéphane d’Ascoli. All slides and notebooks available online (@GoogleColab compatible). pic.twitter.

Read Full Story