Receive the Data Science Weekly Newsletter every Thursday
Easy to unsubscribe at any time. Your e-mail address is safe.
Data Science Weekly Newsletter
May 29, 2014
The Flaw Lurking In Every Deep Neural Net
A recent paper "Intriguing properties of neural networks" by Christian Szegedy, Wojciech Zaremba, Ilya Sutskever, Joan Bruna, Dumitru Erhan, Ian Goodfellow and Rob Fergus, a team that includes authors from Google's deep learning research project outlines two pieces of news about the way neural networks behave that run counter to what we believed - and one of them is frankly astonishing. I'm going to tell you about both...
What Does a Neural Network Actually Do?
To gain an intuitive understanding of what a learning algorithm does, I usually like to think about its representational power, as this provides insight into what can, if not necessarily what does, happen inside the algorithm to solve a given problem. I will do this here for the case of multilayer perceptrons. By the end of this informal discussion I hope to provide an intuitive picture of the surprisingly simple representations that NNs encode...
Algorithmic Tagging of HackerNews (or any other site)
Part of making algorithms more discoverable is creating meta-data tags to classify them. Often sites will allow users to pick their own tags but what if the content had already been generated? This is the problem we faced when trying to tag all the algorithms in our API. Each algorithm had a description page and we believed that using some simple machine learning algorithms already in our API we could generate tags for each one...
Data Scientist - Yodle, New York NY
We are currently seeking a talented Data Scientist for our team. The ideal candidate possesses strong quantitative abilities and the capacity to express those ideas in code. This person must enjoy math and programming and be capable of using them both in a practical setting. The successful candidate will be team oriented and feel comfortable with the dynamic pace of an Internet startup, participating in all phases of product development from research to implementation and maintenance....
Understanding Machine Learning: From Theory to Algorithms
"This is a timely text on the mathematical foundations of machine learning, providing a treatment that is both deep and broad, not only rigorous but also with intuition and insight. It presents a wide range of classic, fundamental algorithmic and analysis techniques as well as cutting-edge research directions. This is a great book for anyone interested in the mathematical and computational underpinnings of this important and fascinating field."
- Avrim Blum, Carnegie Mellon University, Editoral Review