Data Science Weekly Newsletter - Issue 183

Issue #183

May 25 2017

Editor Picks
  • New paint colors invented by neural network
    So if you’ve ever picked out paint, you know that every infinitesimally different shade of blue, beige, and gray has its own descriptive, attractive name. Tuscan sunrise, blushing pear, Tradewind, etc… There are in fact people who invent these names for a living. But given that the human eye can see millions of distinct colors, sooner or later we’re going to run out of good names. Can AI help?...
  • Predicting Lung Cancer: Solution Write-up
    The Data Science Bowl is an annual data science competition hosted by Kaggle. In this year’s edition the goal was to detect lung cancer based on CT scans of the chest from people diagnosed with cancer within a year. To tackle this challenge, we formed a mixed team of machine learning savvy people of which none had specific knowledge about medical image analysis or cancer prediction. Hence, the competition was both a nobel challenge and a good learning experience for us. The competition just finished and our team Deep Breath finished 9th! In this post, we explain our approach...

A Message from this week's Sponsor:


  • Harness the business power of big data.

    How far could you go with the right experience and education? Find out. At Capitol Technology University. Earn your PhD Management & Decision Sciences — in as little as three years — in convenient online classes. Banking, healthcare, energy and business all rely on insightful analysis. And business analytics spending will grow to $89.6 billion in 2018. This is a tremendous opportunity — and Capitol’s PhD program will prepare you for it. Learn more now.


Data Science Articles & Videos

  • A Day in the Life of Americans
    So again I looked at microdata from the American Time Use Survey from 2014, which asked thousands of people what they did during a 24-hour period. I used the data to simulate a single day for 1,000 Americans representative of the population — to the minute. More specifically, I tabulated transition probabilities for one activity to the other, such as from work to traveling, for every minute of the day. That provided 1,440 transition matrices, which let me model a day as a time-varying Markov chain. The simulations below come from this model, and it’s kind of mesmerizing...
  • Using Machine Learning to Explore Neural Network Architecture
    To make this process of designing machine learning models much more accessible, we’ve been exploring ways to automate the design of machine learning models. Among many algorithms we’ve studied, evolutionary algorithms [1] and reinforcement learning algorithms [2] have shown great promise. But in this blog post, we’ll focus on our reinforcement learning approach and the early results we’ve gotten so far...
  • First In-Depth Look at Google's New Second Generation TPU
    This morning at the Google’s I/O event, the company stole Nvidia’s recent Volta GPU thunder by releasing details about its second-generation tensor processing unit (TPU), which will manage both training and inference in a rather staggering 180 teraflops system board, complete with custom network to lash several together into “TPU pods” that can deliver Top 500-class supercomputing might at up to 11.5 petaflops of peak performance...
  • Automated Machine Learning (AML)  — 
    A Paradigm Shift That Accelerates Data Scientist Productivity @ Airbnb

    The scope of AML is ambitious, however, is it really effective? The answer is it depends on how you use it. Our view is that it is difficult to perform wholesale replacement of a data scientist with an AML framework, because most machine learning problems require domain knowledge and human judgement to set up correctly. Also, we have found AML tools to be most useful for regression and classification problems involving tabular datasets, however the state of this area is quickly advancing. In summary, we believe that in certain cases AML can vastly increase a data scientist’s productivity, often by an order of magnitude...
  • A new algorithm for finding a visual center of a polygon
    We came up with a neat little algorithm that may be useful for placing labels and tooltips on polygons, accompanied by a JavaScript library. It’s now going to be used in Mapbox GL and Mapbox Studio. Let’s see how it works...
  • Who Owns England?
    Who owns land is one of England's most closely-guarded secrets. This map is a first attempt to display major landowners in England, combining public data with Freedom of Information requests...
  • Get Up To Speed Fast As A Junior Data Scientist
    You are a new junior data scientist and you want to get started the right way. You want to make sure you don't make the same mistakes others have made early in their data scientist careers because you want to prove to your employers that they made the right choice. As such, you need to figure out how to get up to speed as fast as possible...


  • Deep Learning Engineer - New Relic - San Francisco, CA

    New Relic is a leading digital intelligence company, delivering full-stack visibility and analytics with more than 14,000 paid business accounts. Our Platform provides actionable insights to drive digital business results.Every minute, New Relic collects over 1.37 billion data points from computers, phones, browsers, and applications all over the world. To handle this incredible influx, we built massively scalable systems capable of ingesting, analyzing, and storing this data.

    We’re looking for talented deep learning engineers to join us in our quest to analyze this data and solve immediate, real world challenges facing modern digital businesses. If you’re someone who lives and breathes everything from logistic regression to LTSMs and CNNs, we’d love to hear from you!

    Join us and apply your expertise in machine learning to solve our hardest problems. You’ll build valuable new products and features, and take a significant role in shaping future product and technology directions at New Relic....

Training & Resources

  • Deep learning for natural language processing, Part 1
    Natural language processing is yet another field that underwent a small revolution thanks to the second coming of artificial neural networks. Let’s just briefly discuss two advances in the natural language processing toolbox made thanks to artificial neural networks and deep learning techniques...



P.S. Looking to hire a Data Scientist? Find an awesome one among our readers! Email us for details on how to post your job :) - All the best, Hannah & Sebastian
Sign up to receive the Data Science Weekly Newsletter every Thursday

Easy to unsubscribe. No spam — we keep your email safe and do not share it.