Since the last Nuit Blanche in Review ( October 2014 ), we landed on a comet (ESA's 28 minutes of Terror : Philae's Landing on 67P, Yes, We Can Locate Philae and Here Is How ...). I also co-hosted the Paris Machine Learning #3 Season 2: Building a Data Science Team, Opinion Mining, Word2Vec, Kaggle and noticed some old way of performing robust PCA (The first ever Robust PCA applied in Imaging). The Sparse- and Low-Rank Approximation Wiki was featured and I came up with one Sunday Morning Insight:

We also had quite a few -to say the least) implementations:

- Matrix Completion on Graphs - implementation -
- Spectral Clustering of Graphs with the Bethe Hessian - implementation -
- CKN: Convolutional Kernel Networks - implementation -
- Compressive Sensing Reconstruction Solvers in Julia
- LRSLibrary : 64 Algorithms for Low-Rank and Sparse for Background Modeling and Subtraction in Videos
- MatchLift : Near-Optimal Joint Object Matching via Convex Relaxation - implementation -
- Zero SR1 quasi-Newton method - implementation -
- CVXPY: A Python-Embedded Modeling Language for Convex Optimization Problems - implementation
- Convex Optimization in Julia

Focused entries

- Accelerated NMR spectroscopy with low-rank reconstruction
- An ODE for solvers, a Group approach to regularizers, Sketching for Kernels and Linear Algebra, Johnson-Lindenstrauss is optimal, Coresets of Streaming Data
- Convex Optimization for Big Data
- Provable Bounds for Learning Some Deep Representations
- Nonlinear Information-Theoretic Compressive Measurement Design
- Riemannian Pursuit for Big Matrix Recovery
- Sparse Reinforcement Learning via Convex Optimization
- How to Scale Up Kernel Methods to Be As Good As Deep Neural Nets
- A Randomized Algorithm for CCA
- 3D Shape Reconstruction from 2D Landmarks: A Convex Formulation
- Neural Word Embeddings as Implicit Matrix Factorization
- Stable Autoencoding: A Flexible Framework for Regularized Low-Rank Matrix Estimation
- Nuclear Norm Minimization via Active Subspace Selection
- Randomized Interpolative Decomposition of Separated Representations
- Dimensionality Reduction for k-Means Clustering and Low Rank Approximation

Job

- Two Postdoc positions ( KU Leuven ) ERC Advanced Grant A-DATADRIVE-B for 1-year, extendable at ESAT-STADIUS
- CS Internship in France: Compatibilité de l'apprentissage par représentations parcimonieuses et de la compression avec pertes / Compatibility between sparse machine learning and lossy compression
- Upcoming CSJobs and more ....

Meetings / Slides / Videos / Cfps

- NIPS 2014 proceedings are out, what paper did you fancy ?
- Saturday Morning Videos: Big data and Data science: LIX colloquium 2014
- Saturday Morning Videos: Machine Learning Summer School, Iceland 2014
- Slides : FOCS 2014 Workshop on The Sparse Fourier Transform: Theory and Applications
- Call for Papers : SPARS 2015 – Cambridge UK – 6-9 July 2015

Books

- Book: Sparse Modeling for Image and Vision Processing
- Book: Regularization, Optimization, Kernels, and Support Vector Machines

Other:

- Saturday Morning Video: Yielding of a Yogurt under constant stress
- The researcher's article : a three-part adventure / Le chercheur et son article : une aventure en trois actes

Four image NAVCAM mosaic comprising images taken on 20 November. Credits: ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0

**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.