## Finds in arxiv, october

This is duplication of my ongoing G+ series of post on interesting for me papers in arxiv. Older posts are not here but in my G+ thread.

Finds in #arxiv :

*Optimization, numerical & convex, ML*

The Linearized Bregman Method via Split Feasibility Problems: Analysis and Generalizations

Reformulation of Split Bregman/ ADMM as split feasibility problem and algorithm/convergence for generalized split feasibility by Bregman projection. This general formulation include both Split Bregman and Kaczmarz (my comment – randomized Kaczmarz seems could be here too)

http://arxiv.org/abs/1309.2094

Stochastic gradient descent and the randomized Kaczmarz algorithm

Hybrid of randomized Kaczmarz and stochastic gradient descent – into my “to read” pile

http://arxiv.org/abs/1310.5715

Trust–Region Problems with Linear Inequality Constraints: Exact SDP Relaxation, Global Optimality and Robust Optimization

“Extended” trust region for linear inequalities constrains

http://arxiv.org/abs/1309.3000

Conic Geometric Programming

Unifing framwork for conic and geometric programming

http://arxiv.org/abs/1310.0899

http://en.wikipedia.org/wiki/Geometric_programming

http://en.wikipedia.org/wiki/Conic_programming

Gauge optimization, duality, and applications

Another big paper about different, not Lagrange duality, introduced by Freund (1987)

http://arxiv.org/abs/1310.2639

Color Bregman TV

mu parameters in split bregman made adaptive, to exploit coherence of edges in different color channels

http://arxiv.org/abs/1310.3146

Iteration Complexity Analysis of Block Coordinate Descent Methods

Some convergence analysis for BCD and projected gradient BCD

http://arxiv.org/abs/1310.6957

Successive Nonnegative Projection Algorithm for Robust Nonnegative Blind Source Separation

Nonnegative matrix factorization

http://arxiv.org/abs/1310.7529

Scaling SVM and Least Absolute Deviations via Exact Data Reduction

SVN for large-scale problems

http://arxiv.org/abs/1310.7048

Image Restoration using Total Variation with Overlapping Group Sparsity

While title is promising I have doubt about that paper. The method authors suggest is equivalent to adding averaging filter to TV-L1 under L1 norm. There is no comparison to just applying TV-L1 and smoothing filter interchangeably.The method author suggest is very costly, and using median filter instead of averaging would cost the same while obviously more robust.

http://arxiv.org/abs/1310.3447

*Deep learning*

Deep and Wide Multiscale Recursive Networks for Robust Image Labeling

_Open source_ matlab/c package coming soon(not yet available)

http://arxiv.org/abs/1310.0354

Improvements to deep convolutional neural networks for LVCSR

convolutional networks, droput for speech recognition,

http://arxiv.org/abs/1309.1501v1

DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition

Already discussed on G+ – open source framework in “learn one use everywhere” stile. Learning done off-line on GPU using ConvNet, and recognition is online in pure python.

http://arxiv.org/abs/1310.1531

Statistical mechanics of complex neural systems and high dimensional data

Big textbook-like overview paper on statistical mechanics of learning. I’ve put it in my “to read” pile.

http://arxiv.org/abs/1301.7115

Randomized co-training: from cortical neurons to machine learning and back again

“Selectron” instead of perception – neurons are “specializing” with weights.

http://arxiv.org/abs/1310.6536

Provable Bounds for Learning Some Deep Representations

http://arxiv.org/abs/1310.6343

Citation:”The current paper presents both an interesting family of denoising autoencoders as

*Computer vision*

Online Unsupervised Feature Learning for Visual Tracking

Sparse representation, overcomplete dictionary

http://arxiv.org/abs/1310.1690

From Shading to Local Shape

Shape restoration from local shading – could be very useful in low-feature environment.

http://arxiv.org/abs/1310.2916

Fast 3D Salient Region Detection in Medical Images using GPUs

Finding interest point in 3D images

http://arxiv.org/abs/1310.6736

Object Recognition System Design in Computer Vision: a Universal Approach

Grid-based universal framework for object detection/classification

http://arxiv.org/abs/1310.7170

Gaming :)

Lyapunov-based Low-thrust Optimal Orbit Transfer: An approach in Cartesian coordinates

For space sim enthusiast :)

http://arxiv.org/abs/1310.4201

## ConvNet for windows

I have seen an excellent wlakthrough on building Alex Krizhevsky’s cuda-convnet for windows, but difference in configuration and installed packages could be tiresome. So here is complete build of convnet for windows 64:

https://github.com/s271/win_convnet

It require having CUDA compute capability 2.0 or better GPU of cause, Windows 64bit, Visual Studio 64bits and Python 64bit with NumPy installed. The rest of libs and dlls are precomplied. In building it I’ve followed instructions by Yalong Bai (Wyvernbai) from http://www.asiteof.me/archives/50.

Read Readme.md before installation – you may (or may not) require PYTHONPATH environmental variable set.

On side note I’ve used WinPython for both libraries and running the package. WinPython seems a very nice package, which include Spyder IDE. I have some problems with Spyder though – sometimes it behave erratically during debugging/running the code. Could be my inexperience with Spyder though. Another nice package – PythonXY – regretfully can not be used – it has only 32 bit version and will not compile with/run 64 bit modules.