Frequently Asked Questions
...about this web site
1. When will my submissions show up on the web site?
Submissions of papers, URLs, events, etc. to this web site automatically get collected during the week. At the end of each week, an email of changes is sent to the administrators. They then have one week's time to ensure that everything is OK, and to fix any remaining HTML bugs. At the end of that week, the changes go online. This means that in the worst case, it will take two weeks before your changes will show up online.
2. How is it decided whether a submitted paper will actually be put on the web site?
This web site tries to be as open as possible. At present, we do not review any papers - everything that's within the scope of the site will be published.
3. How about the quality of the information on this web site? Is this some kind of an electronic journal?
This is not a scholarly journal, it is a mere source for the fast dissemination of information. As we do not review the submissions, the editorial board takes no responsibility whatsoever for the content, or the quality of the papers or anything else that appears on this site. Also, we do not take any responsibility for whatever happens when you use any of the algorithms.
4. Why are there currently two websites on Support Vector Machines?
The GMD site (svm.first.gmd.de) is the old website and will be shut down in the near future in such a way that links pointing to it should result in a redirect to www.kernel-machines.org. The time when this may happen depends on the current maintainers of the website (Bernhard Schölkopf and Alex Smola are no longer in control of the site). For information about research at GMD Berlin on kernels look here.
...about SVMs and kernel methods
1. I know nothing about SVMs. What should I read to learn about them fast?
There are several introductions, review papers, and books. As usual, there is a trade-off between how much time you want to invest, and how much you will get out of it. The following list is ordered by increasing time. On the log(time) domain, the increase in effort should be pretty much linear. ;)
(a) The introduction of the book Advances in Kernel Methods - Support Vector Learning or the high level overview of Hearst et al. from IEEE Intelligent Systems
(b) The tutorial papers of Burges (SVM pattern recognition) or Smola and Schoelkopf (SVM regression estimation)
(c) The small book of Vapnik, published by Springer (1995, or, in second edition, 1999), or the one of Cristianini and Shawe-Taylor (2000).
Alternatives that can be downloaded free of charge are PhD theses on SVMs, such as the ones of Schoelkopf (1997), Smola (1998), Herbrich (2000, to be finished soon).
(d) The collection of papers presented at the NIPS workshop on SVMs and kernel methods, from 1997 (Advances in Kernel Methods, MIT Press, 1999), 1998 (Advances in Large Margin Classifiers, MIT Press, 2000 - to appear), or 1999 (these will appear in a special issue of the journal Machine Learning at some point of time in the future).
(e) The long book of Vapnik, the bible of statistical learning theory (Wiley, 1998).
2. I know nothing about Gaussian processes. What should I read to learn about them fast?
We have included a link to David MacKay's comprehensive tutorial on this web site.
For a point of view that emphasizes commonalities with SVMs, you might want to take a look at the ICANN'99 tutorial about Kernel Methods: Prediction with Support Vector Machines and Gaussian Processes.