Around the Blogs in 78 Summer Hours

Some news first: Stan Osher received the Gauss Prize ( the other prizes are here but a good write-up is here). Terry has a more in-depth write-up on the Fields medalists. Here are a few blog entries of interest: 

Danny let me know about two Sparse K-means papers on his blog. There is also Interesting paper from Dataiku about WCSD 2014.

Masaaki:

• Sparse Packetized Predictive Control for Networked Control over Erasure Channels
• L1 Control Theoretic Smoothing Splines

Property Testing Review (The latest in property testing and sublinear time algorithms) Eric Blais:

• News for July 2014 on Sublinear Algorithms for Dense r-CSPs and Lp testing

Afonso:

• An interesting problem by Mallat and Zeitouni

Bob:

• Subspaces of Comparative Experiments
• EUSIPCO 2014 papers announced
• How reproducibility tipped the scale toward article acceptance

Dustin:

• Introduction to fast matrix multiplication
• Derandomizing restricted isometries via the Legendre symbol

Dirk:

• Remembering the Douglas-Rachford iteration.

Tianyi:

• Multi-task Copula – A semiparametric joint prediction model for multiple outputs with sparse graph structure

John

• Vowpal Wabbit for Fast Learning

In Interdisciplinary research and the intellectual richness of data analysis, Suresh wrote:

What topics do you need a more than passing familiarity with to do data analysis ? (when I say "do", I don't mean the ability to run and understand standard methods. I mean a level of facility where you understand the underlying principles behind different methods, and are capable of designing new methods yourself: so roughly at the level of a graduating Ph.D student)

At the very least, the list of topics includes:

• Linear algebra (and a sound understanding of matrices, transformations, projections, eigenspace analysis, matrix decompositions, and the like)

• Probability (including slightly more advanced notions like conditional probabilities and conditional expectation, mutual information, and entropy, sampling)

• Basic statistics (estimators, distributions, hypothesis testing, Bayesian analysis)

• Functional analysis (vector spaces, norms, Hilbert spaces and so on)

• Optimization (linear to convex, and beyond. Also some understanding of gradient descent and methods beyond to do optimization)

• Basic algorithm design (design primitives, randomization, approximation)

• Scalable algorithm design (streaming, distributed, sketching)

• Limits of computing (a passing familiarity with basic complexity theory, just to know what you can't expect to do)

• High dimensional geometry (and topology)

• Core ML (basic tools in learning, some smattering of learning theory)

That's a lot of knowledge !! And much of this is still evolving, and is not always in textbook form or in any way easy to assimilate. What's worse is that different areas might discuss the same topics with completely different vocabulary. 

Yes, it is indeed very much evolving and Nuit Blanche tries to follow that trend, but it is not straightforward!

Aleks

• Data & Visualization Tools to Track Ebola

Randy

• Terahertz Chip Identifies Short Strands of DNA, well come back to that.

Gabriel:

• So you wanna try deep learning

Boris

• convolutional learnings: things i learned by implementing convolutional neural nets

Harri Edwards

• Theano, Autoencoders and MNIST

Google Research

• Doing Data Science with coLaboratory

Christian

• Ebola virus [and Mr. Bayes]
• ABC model choice by random forests [guest post]

Alex

• Julia, once more ( which led me to this interesting site Learn X in Y minutes)

Kyle

• Introduction to Gaussian Processes

See invisible  (with a Webcam/C++ implementation of the recent MIT magnification work )
 

Here is an offer from theWinnower:

Science bloggers: assign a DOI to your writings, write in a community, empower your work. Sign up at http://t.co/xoUyb2wUuy
— theWinnower (@theWinnower) August 13, 2014

In the meantime, since July's Nuit Blanche in Review, Nuit Blanche featured the following entries:

• DDRS : Nonlinear Dimensionality Reduction of Data by Deep Distributed Random Samplings - implementation -
• Reproducible Research: MRI Data
• From Denoising to Compressed Sensing - implementation -
• Deep Learning and Convolutional Kernel Networks
• Sunday Morning Insight: Why Kaggle Changes Everything
• Exponential decay of reconstruction error from binary measurements of sparse signals - implementation -
• sFFT-DT: Sparse Fast Fourier Transform for Exactly and Generally K-Sparse Signals by Downsampling and Sparse Recovery
• Arrival at 67P
• WiTrack and RTI Goes Wild: Indoor and Outdoor Radio Tomographic Imaging
• Oldies but Goodies
• The Visual Microphone: Passive Recovery of Sound from Video
• Review: Low Complexity Regularization of Linear Inverse Problems
• The application of Compressed Sensing for Longitudinal MRI / Multichannel Compressive Sensing MRI Using Noiselet Encoding
• Kernel nonnegative matrix factorization without the curse of the pre-image
• Sunday Morning Insight: Physics Driven Sensor Design ?
• January-July 2014: Seven Months of Reproducible Research
• Nuit Blanche in Review (July 2014)

Image Credit: NASA/JPL/Space Science Institute,Full-Res: W00088926.jpg

W00088926.jpg was taken on August 05, 2014 and received on Earth August 07, 2014. The camera was pointing toward SATURN at approximately 1,742,745 miles (2,804,676 kilometers) away, and the image was taken using the CB2 and CL2 filters.

Join the CompressiveSensing subreddit or the Google+ Community and post there !
Liked this entry ? subscribe to Nuit Blanche's feed, there's more where that came from. You can also subscribe to Nuit Blanche by Email, explore the Big Picture in Compressive Sensing or the Matrix Factorization Jungle and join the conversations on compressive sensing, advanced matrix factorization and calibration issues on Linkedin.