Navigation


RSS Feed: Blog



Showing Items 11-16 of 16 on page 2 of 2: Previous 1 2

Shogun got accepted at Google Summer of Code 2013

We are happy to announce that the shogun machine learning toolbox will participate in this years google summer of code :-)

SHOGUN is designed for unified large-scale learning for a broad range of feature types and learning settings, like classification, regression, or exploratory data analysis.

In case you are a talented student interested in implementing and learning about machine learning algorithms - we are looking for you!

We have collected a number of ideas [1] that we consider worthwhile implementing. And don't forget to apply [2]!

[1] http://shogun-toolbox.org/page/Events/gsoc2013_ideas

[2] http://google-melange.appspot.com/gsoc/org/google/gsoc2013/shogun


CfP: Shogun Machine Learning Workshop, July 12-14, Berlin, Germany

CALL FOR PARTICIPATION: Shogun Machine Learning Workshop, Berlin, Germany, July 12-14, 2013

Data Science, Big-Data are omnipresent terms documenting the need for automated tools to analyze the ever growing wealth of data. To this end we invite practitioners, researchers and students to participate in the first Shogun machine learning workshop. While the workshop is centered around the development and use of the shogun machine learning toolbox, it will also feature general machine learning subjects.

General Information

The workshop will include:
  • A general introduction to machine learning held by Gunnar Raetsch.
  • Introductory talks about e.g. Dimension reduction techniques, Kernel-statistical testing, Gaussian Processes, Structured Output learning.
  • Contributed talks and a poster session, and a poster-spotlight.
  • A discussion panel
  • A hands on session on July 13-14

Do not miss the chance to familiarize yourself with the shogun machine learning toolbox for solving various data analysis tasks and to talk to their authors and contributors. The program of the workshop will cover from basic to advanced topics in machine learning and how to approach them using Shogun, which makes it suitable for anyone, no matter if you are a senior researcher or practitioner with many year's of experience, or a junior student willing to discover much more. Interested?

A tentative schedule is available at http://shogun-toolbox.org/page/Events/workshop2013_program.

Call for contributions

The organizing committee is seeking workshop contributions. The committee will select several submitted contributions for 15-minute talks and poster presentations. The accepted contributions will also be published on the workshop web site.

Amongst other topics, we encourage submission that

  • are applications / publications utilizing Shogun
  • are highly relevant to practitioners in the field
  • are of broad general interest
  • are extensions to Shogun

Submission Guidelines

Send an abstract of your talk/contribution to shogun-workshop2013@shogun-toolbox.org before June 1. Notifications will be given on June 7.

Registration

Workshop registration is free of charge. However, only a limited number of seats is available. First-come, first-served! Register by filling out the registration form.

Location and Timeline

The main workshop will take place at c-base Berlin (http://c-base.org/, https://en.wikipedia.org/wiki/C-base) on July 12. It is followed by additional 2-day hands-on sessions held at TU Berlin on July 13-14.

About the Shogun machine learning toolbox

Shogun is designed for unified large-scale learning for a broad range of feature types and learning settings, like classification, regression, or explorative data analysis. Further information is available at http://www.shogun-toolbox.org.


Shogun Student Applications Statistics for Google Summer of Code 2013

Almost a month has passed since SHOGUN has been accepted for Google Summer of Code 2013. Student application deadline was today (May 6) and shogun received 57 proposals from 52 students. This is quite some increase compared to 2012 (48 applications from 38 students). What is interesting though is that it didn't look that good in the very beginning (see the figure below):

Comparing this to 2012, this curve is much more flat in the beginning but exponentially increasing towards the end. Why is that? We didn't change the way we engaged with students (even though we tried to improve the instructions and added lots of entrance tagged tasks to github issues). We still require patches to be submitted to even be considered. So it is similarly tough to get into gsoc 2013 with us as it was in the previous year.

What is interesting though is that various organizations complained about a slow uptake in the beginning. And it turns out that google did limit the number of student applications from 20 (last year) to 5 (in 2013). This might explain the shape of the curve: Students are more cautious to apply but once the deadline is near the apply to the maximum of 5 to improve their chances. This goes hand-in-hand with the observation that the quality of newly submitted student applications tends to decrease towards the deadline.

So did this new limit hurt? To the contrary! In the end the quality of proposals increased a lot and we were able to even way before the student application deadline start to score/rank students. We are happy to have many very strong candidates this year again. Lets hope we get enough slots to accommodate all of the excellent students and then lets start the fun :)


Shogun Toolbox Days 2013 Program and Updates

Dear all,

we are excited that the first Shogun workshop, July 12-14 in Berlin, is getting closer. Thanks to all the people that signed up -- we are sure it will be a packed and inspiring weekend!

We have finalized the schedule for Friday, July 12, taking place at the C-Base (see description below [1]). After an intro where everyone gets to know each other and where we introduce ourselves, Shogun, and Machine Learning in general, there will be some tutorials by Shogun developers. In addition, we will have discussions about various topics and various coffee breaks and lunch. Finally, we will enjoy a summer's evening in Berlin.

On Saturday and Sunday, July 13-14 there will be hands on sessions at the Technical University Berlin [2], where developers are around for more close up discussions and practical guidance. Bring your Laptop if you want to try things.

See the final schedule for more details [1]. We plan to do video recordings of all lectures and will have a live stream [3].

See you there! The Shogun-Team


Shogun goes cloud

Shogun goes cloud. Try out http://cloud.shogun-toolbox.org to interactively play with machine learning algorithms or try any of the interactive demos.

The SHOGUN Machine Learning Toolbox is a collection of algorithms designed for unified learning for a broad range of feature types and learning settings, like classification, regression, or explorative data analysis. For more information visit http://www.shogun-toolbox.org.


Shogun Toolbox Version 3.0 released!

Dear all,

we are proud to announce the 3.0 release of the Shogun Machine-Learning Toolbox. This release features the incredible projects of our 8 hard-working Google Summer of Code students. In addition, you get other cool new features as well as lots of internal improvements, bugfixes, and documentation improvements. To speak in numbers, we got more than 2000 commits changing almost 400000 lines in more than 7000 files and increased the number of unit tests from 50 to 600. This is the largest release that Shogun ever had! Please visit http://shogun-toolbox.org/ to obtain Shogun.

News

Here is a brief description of what is new, starting with the GSoC projects, which deserve most fame:

  • Gaussian Process classification by Roman Votjakov
  • Structured Output Learning of graph models by Shell Hu
  • Estimators for log-determinants of large sparse matrices by Soumyajit De
  • Feature Hashing and random kitchen sinks by Evangelos Anagnostopoulos
  • Independent Component Analysis by Kevin Hughes
  • A web-based demo framework by Liu Zhengyang
  • Metric learning with large margin nearest neighbours by Fernando Iglesias
  • Native support for various popular file formats by Evgeniy Andreev

Screenshots

Everyone likes screenshots. Well, we have got something better! All of the above projects (and more) are now documented in the form of IPython notebooks, combining machine learning fundamentals, code, and plots. Those are a great looking way that we chose to document our framework from now on. Have a look at them and feel free to submit your use case as a notebook!

FGM.html GMM.html HashedDocDotFeatures.html LMNN.html SupportVectorMachines.html Tapkee.html bss_audio.html bss_image.html ecg_sep.html gaussian_processes.html logdet.html mmd_two_sample_testing.html

The web-demo framework has been integrated into our website, go check them out.

Other changes

We finally moved to the Shogun build process to CMake. Through GSoC, added a general clone and equals methods to all Shogun objects, and added automagic unit-testing for serialisation and clone/equals for all classes. Other new features include multiclass LDA, and probability outputs for multiclass SVMs. For the full list, see the NEWS.

Workshop Videos and slides

In case you missed the first Shogun workshop that we organised in Berlin last July, all of the talks have been put online.

Shogun in the Cloud

As setting up the right environment for shogun and installing it was always one of the biggest problems for the users (hence the switching to CMake), we have created a sandbox where you can try out shogun on your own without installing shogun on your system! Basically it's a web-service which give you access to your own ipython notebook server with all the shogun notebooks. Of course you are more than welcome to create and share your own notebooks using this service! *NOTE*: This is a courtesy service created by Shogun Toolbox developers, hence if you like it please consider some form of donation to the project so that we can keep up this service running for you. Try shogun in the cloud.

Thanks

The release has been made possible by the hard work of all of our GSoC students, see list above. Thanks also to Thoralf Klein and Björn Esser for the load of great contributions. Last but not least, thanks to all the people who use Shogun and provide feedback.

Sören Sonnenburg on behalf of the Shogun team (+ Viktor Gal, Sergey Lisitsyn, Heiko Strathmann and Fernando Iglesias)


Showing Items 11-16 of 16 on page 2 of 2: Previous 1 2

About Me

Sören Sonnenburg Dr. Sören Sonnenburg
Machine Learning / Bioinformatics Research Scientist
Berlin, Germany

News


Links