Category Archives: University

A Long Journey: Acceler8 and TBB

Posted by on November 16, 2011 No comments

A month has passed with Intel's Acceler8 competition and it has finally come to an end. It's been a long way from implementing the first sequential algorithm to having a fully-fledged parallel version.

I have never worked with Intel's Threading Building Block library before and it was a nice opportunity to examine it, since it offered a better abstraction than OpenMP or pthreads.

The documentation is very good and you quickly learn how to work with the library. One of the caveats is that I didn't have to use a low-level synchronization construct once in the development and everything worked fine without any race conditions or similar . The parallel_* functions (eg parallel_for, parallel_reduce, and parallel_scan) together with icc's C++0x support (lambda functions) allowed for very concise code and little programming overhead.

The implementation builds on Kadane's algorithm for the two dimensional case using prefix sums. One implementation that gets across the basic idea can be found here. Mine is similar and I simply parallelized as much as possible.

As you can see the outer two loops iterate over a two-dimensional range that is pretty much an upper right triangle of the whole possible domain. For this I've implemented a custom range that allows for better load balancing. A range in TBB defines an iteration range of any kind and supports a split operation that is used internally by the task scheduler to distribute the range dynamically on multiple threads as it sees fit.

Last but not least I came up with a way to parallelize the 1D part of Kadane's algorithm that is being used by splitting the column range into linear subranges and merging the subsolutions into one solution, ie a classical divide and conquer approach.

Because it's the most abstract yet interesting part of our implementation, I'm going to go into more detail here. :-)

How can you find the maximum subarray of a 1D array, if you know the maximum subarray of the two "halves" (they don't have to be split evenly)?
Well, you don't, you need more information.

We calculate the following information for each chunk:

  • maximum subarray that starts at the beginning of the chunk
  • maximum subarray that ends at the end of the chunk
  • total sum
  • maximum subarray

It's easy to figure out how to merge these values for two neighboring chunks into the values of the merged chunk.
The merged maximum subarray that starts at the beginning of the merged chunk is either said value for the left chunk or the total sum of the left chunk + that value of the right chunk.
You can figure out how it works for the maximum subarray that ends at the end of the merged chunk :-)
The maximum subarray is just the biggest of all merged values or the left maximum subarray or right one.

Using this idea you can use a simple parallel_reduce to parallelize Kadane's algorithm.

Of course, there is some overhead but for the right problem sizes this will be faster than the sequential algorithm as always.

Two more take-aways:

  • Always try to use language features like templates or lambda expressions to reduce duplicate code or make the code more concise.
  • Write unit tests. I have used googletest which is a very small but very capable library, and it has spared me a lot of debugging trouble.

Cheers :-)

Reading Nonfiction Books Quickly

Posted by on November 2, 2011 No comments

I like to read books and I also like to read nonfiction books and in particular nonfiction books that I do not agree with.

Speed Reading

For this I've decided to look into speed reading. It is an umbrella term for methods that increase your reading speed while keeping a satisfactory comprehension level. A usual reading speed is 200-300 wpm (words per minute). Speed reading advertises increasing your speed to > 600 wpm.

I've searched around for a bit and read lots of blog posts and articles and here are my favorites:

http://www.fourhourworkweek.com/blog/2009/07/30/speed-reading-and-accelerated-learning/ is a good introduction to the main techniques of speed-reading. It also includes some exercises and is a short read.

http://noahfleming.com/blog/speed-read-like-rain-man-75-increased-reading-speed-in-20-minutes describes the author's personal experience with fourhourworkweek's article.

There is also speed-reading software available. The idea is your eye movement is the main hindrance to reading really fast, so this software displays the text for you to read in groups of several words in the center of the screen. This way you can read everything at once without having to move your eyes at all.
As crazy as it sounds, it works to a great degree.

http://www.spreeder.com/ is  an online reader that works this way. Give it a try! You can make it display the introduction text at 600 wpm or 800 wpm and see how you'll understand - you'll be surprised!

'The Speed Reading Workbook' is what I've been using to practise it. It's okay written and has plenty of exercises and includes timing and evaluation sheets, which are really useful for measuring your progress.

A good summary of many concepts can be found in the PowerPoint Presentation 'Double your Reading Speed in 25 Minutes'.

About Reading in General

http://www.matthewcornell.org/blog/2006/2/26/how-to-read-a-lot-of-books-in-a-short-time.html is a good read and http://pne.people.si.umich.edu/PDF/howtoread.pdf has many good suggestions about how to increase your reading productivity.

Own Expierence

I've been exercising with the workbook now and then for the past weeks and I've become faster but this is still an ongoing process for me, so I'm going to blog about it later (or rather update this post).

For what it's worth: my reading speed was 300 wpm and now it is around 550 wpm :-)

Presentation Advice

Posted by on August 14, 2011 No comments

So.. the summer term has passed and I'm wrapping things up and one of the final things I want to get out of my notebooks into this blog are some notes I took for feedback during my Hauptseminar in my maths studies.

Everybody had to do a 60 minutes presentation and the audience was asked to take notes about the presentation style to give feedback later (in addition to paying attention to the presentations).

There were 4 blackboards in the room and a beamer.

Here are my notes:

  • 4 pages are the most you can do in 60 minutes (ie write down on the blackboard)
  • really think about your audience and what you want them to take away and focus on that in your presentation
  • practice your presentation a few times before your big day
  • if you use formal mathematical symbolism to express an idea, make sure all symbols and variables are properly defined/explained (for it to make sense!)
  • never tell people to "only ask questions at the end because it would disturb your presentation"!
  • focus on important concepts and ideas
  • motivate definitions and formulas - prefer an algorithmic view on things
  • try to use images and graphs to visualize ideas
  • if you can't avoid complex definitions and deductions, use a handout
  • make sure the handout doesn't contain superfluous information which will only confuse readers
  • don't just copy your handout onto the blackboard, if you hand it out at the beginning of the presentation (it's okay, of course, if you do it afterwards)
  • never clean all blackboards at the same time
  • avoid long breaks in your presentation
  • keep on talking all the time, even while cleaning the blackboard to keep people engaged
  • if the blackboards are in front of each other, always start writing on the one on the back
  • if you can use PowerPoint, try to use it for complex/boring formulas or for visualizations, which are impossible to draw by hand
  • on the other hand, make sure, you don't put things into PowerPoint, that would be better written out slowly on the blackboard
  • always listen to criticism politely, but don't take notes of it, while you're in front of your audience
  • don't degrade yourself by apologizing exaggeratedly for mishaps
  • put the sources of referenced material onto the slide the material is used in
  • don't waste time on complicated 'examples' that are too detailed to be of any use for the audience
  • don't let technicalities obscure examples - stick to the important parts of them
  • it's nice if the handout doesn't exactly match the presentation
  • always conclude your presentation with a summary of the main points of your talk
  • try to build intuition for your topic in your audience instead of explaining difficult proofs no one will remember or care for later anyway
  • make sure you have chosen the right priorities for your presentation and trim it down as much as possible
This was it :-)
Cheers,
 Andreas

Books on Numerical Analysis

Posted by on August 1, 2011 No comments

Due my writing an exam on numerical analysis I had the pleasure to look through lots and lots of books on numerical analysis, and here is a list of my favorite ones so far:

  • Afternotes on Numerical Analysis & Afternotes Goes to Graduate School by G. W. Stewart
    Both books are very readable and introduce many of the concepts in a light way that builds an intuitive understanding for them.
    It's possible to read the books cover to cover in a few days and you can learn a lot very quickly.
    They are also quite amusing:

    "In the nineteenth century the Norwegian mathematician Niels Abel showed that no polynomial of degree five could be solved by a finite number of additions, multiplications, divisions, and root extractions. If we had a finite algorithm for finding eigenvalues of general matrices, we could apply it to companion matrices and make a fool out of Abel. Abel was no fool."

  • Numerical Linear Algebra by Trefethen
    another very good book which I haven't used much personally, though
  • Numerical Methods by Kelley
    contains an okay introduction to CG and GMRES.
  • Numerische Mathematik I & Numerische Mathematik II
    are very good books, too. The first volume contains a good introduction to everything but ODEs and PDEs and the second volume consists of a very thorough overview of numerical algorithms for differential equations, including a nice introduction to the general theory of their solvability etc
  • Numerical Analysis by Burden and Faires
    a very nice book that contains lots of visualizations and covers many topics
More to follow soon :-)
Cheers,
 Andreas

ANTLR Stupidity (Warning 209)

Posted by on September 6, 2010 2 comments

I've been playing around with a Java grammar for ANTLR that was supposed to work straight-away but it did not, with very strange warnings and errors, that made it look like ANTLR only supports lexers with a lookahead of 1 character:

warning(209): ...: Multiple token rules can match input such as "'*'": STAR, STAREQ

while STAR matches only '*' and STAREQ only matches '*='. This is a huge w-t-f, especially if you have worked with ANTLR before and didn't have issues with this. This also contradicted all documentation you can find about ANTLR and its lexer rules.

I've spent a considerable amount of time with Google trying to find how to fix it. First I've found lots of posts on the ANTLR mailing list [antlr-interest] from people who had the same issue and no replies to them (really helpful, eh?). People had issues with replacing character ranges with unicode ranges (or rather a huge list of unicode characters), which probably caused the problem in my grammar, too. Others found that ANTLR suddenly behaved as if it only had a one character lookahead, but only if more than 300 lexer rules were used in the grammar.

After searching for a long time and almost giving up on the mini-project I've wanted to use ANTLR for, I've found this post: http://www.antlr.org/pipermail/antlr-interest/2009-September/035954.html (which matches my problem more or less but with additional insight)
and someone even replied (someone being the guy who maintains the C runtime of ANTLR):
http://www.antlr.org/pipermail/antlr-interest/2009-September/035955.html

If you are sure that the messages are not correct and the lexer rules
are not ambiguous, then you probably need to increase the conversion
timeout:

-Xconversiontimeout 30000

if that does not work, then there is a conflict in your rules.

Jim

And that turns out to be the right advice and the remedy to my problems and the problems of lots of other people probably.
However, no warning or error message I encountered mentioned that ANTLR's internal processing actually timed-out and there was no ambiguity in the grammar itself...

This comes to show that any good tool like ANTLR can quickly degrade to a piece of crap and a major source of annoyance, if error and warning messages aren't clear and helpful.

On further investigation, you can trigger warnings that the conversion times out:

internal error: org.antlr.tool.Grammar.createLookaheadDFA(Grammar.java:1279):
    could not even do k=1 for decision 121; reason: timed out (>1ms)

but not consistently. I guess this is a bug - either in ANTLR or in ANTLRWorks... :-|

Light Propagation Volumes

Posted by on July 18, 2010 19 comments

I've finally finished my lab course last week - thanks to my supervisor Matthäus G. Chajdas - you can read his blog here -, it wasn't your usual lab course with work sheets and boring homework, instead I've been allowed to implement a nice paper about a Global Illumination approximation algorithm called (Cascaded) Light Propagation Volumes. It's been developed by Crytek and you can find more information (including some presentations and videos) on their server. (Note: this is an implementation of the I3D paper, not the earlier SIGGRAPH one.)

Sponza scene (direct + indirect lighting w/ occlusion)

Sponza scene (only indirect lighting w/ occlusion)

Sponza scene (ony direct lighting)

Sponza scene (boosted indirect lighting w/ occlusion)

Sponza scene (boosted indirect lighting w/o occlusion)

The algorithm approximates global illumination by rendering the light into a reflective shadow map, injecting it into a volume (using a spherical harmonics representation) and propagates the light flux in this volume (hence the name of the algorithm) and taking into account occlusion as possible extension.

The whole algorithm is physically motivated but corners everywhere, of course, to be more efficient. The paper also contains a few errors and doesn't explain everything needed to implement it in great detail (like eg the solid angles of the side faces), so I've written two documents detailing the mistakes I've found and the additional calculations I've performed.

You can find the mistakes here (including suggested corrections) and the full annotations document here.

Finally I've also uploaded the whole prototype (including my code licensed under the FreeBSD license and the media files) here - it's 68 MB big (and it's been compressed with 7zip with a compression mode that might not be supported by WinZIP. The Sponza model is from Crytek, too. You can download the original model and textures here.
The project uses DirectX 10.1 and by default it won't run in DirectX 10, because it uses a texture format that is deprecated in D3D10 but supported again 10.1 (BGRA). See the comment by FatGarfield for the location that needs to be changed for it work in DX10, too. (However red and blue will be swapped then.)

I haven't implemented cascaded LPVs and I also use only one light/RSM and only inject its depth into the occlusion volume, but the results already look very nice in my opinion.

Stay tuned for more :-)
Cheers,
Andreas

Panorama Stitching

Posted by on July 17, 2010 7 comments

I've finally come around to "clean-up" some old project I've had lying around for a few months and upload it.
I'm talking about some Panorama Stitching code I wrote for our participation in Microsoft's Imagine Cup.
I'm suppressing all memories of it since it was an epic fail, but at least I still have learnt quite a bit about computer vision and image processing - enough to know that it's incredibly hard to come up with stable algorithms and kudos to anyone working in the field.

Here is the code dump: PanoramaStitching.zip

It contains many small projects which usually use multiple pictures as inputs or multiple webcams (depending on code or chosen preprocessor macros).

The most advanced prototype is the SnapshotHomographyConfigurator, which allows you to determine homographies between multiple cameras at once by marking shared points between the images.

Another one which works okay is the PanoramaStitching project. It creates panoramas using spherical or cylindircal projections of the input images. However, it is very sensitive to translations of the viewpoint. It works quite well with optimal/artificial images:

(Note: the small misalignment on the right stems from moving the player position slightly. Usually you use a deghosting algorithm to remove such misalignments.)

I've used OpenCV for image processing and yaml for loading and storing settings (and also rapidxml). OpenCV's C++ wrapper is pretty awesome. It's not perfect but it makes life a lot easier.

Stay tuned for more code/project uploads soon :-)

PS: Here are the links to some papers which proved useful to me (I didn't implement most of them though, and some are implemented in OpenCV already):

Rigid Body Motion

Posted by on May 30, 2010 2 comments

Last week I had to give a presentation about Rigid Body Motion (ie the basics of rigid body physics and some general mechanics).

Here are two versions of my presentation (one with less text and one with more):

Rigid Body Motion Presentation as PPTX; Rigid Body Motion Presentation as PDF
Rigid Body Motion Full Version as PPTM; Rigid Body Motion Full Version as PDF

If you are truly interested in learning about rigid body physics, here are some books/links:

Cheers,
Andreas

A quick note on quantifiers: For almost all and there exist infinitely many

Posted by on March 27, 2010 No comments

It's been a while since my last post and now it's time for a mathematical post:

I'm currently preparing for a math exam (calculus) and I'm thinking it would be nice if there was a way to avoid much of the "for every \epsilon > 0 there is an N \in \mathbb{N}, so that for all n \ge N some property ... holds" stuff you find in textbooks. Some textbooks actually shorten it to "for every \epsilon > 0 some property ... holds for almost all n".
However, I haven't found a quantifier to express this anywhere yet, so I'm proposing to introduce a new one:

\tilde{\forall} x \in M: P(x) should mean "for almost all x \in M P(x) holds", which suggests that there are only finitely many elements for which it does not hold.
One can formalize this as:
\tilde{\forall} x \in M: P(x) \equiv \exists n \in \mathbb{N}: \left | \left \{ x \in M \mid \neg P(x) \right \} \right | \le n

Of course, a new existential quantifier is required then, too (for negation):
\tilde{\exists} x \in M: P(x) stands for "there exist infinitely many x \in M, for which P(x) holds".
And this can be formalized as:
\tilde{\exists} x \in M: P(x) \equiv \forall n \in \mathbb{N}: \left | \left \{ x \in M \mid P(x) \right \} \right | > n

It's easy to see that \neg \left ( \tilde{\forall} x \in M: P(x) \right ) \equiv \tilde{\exists} x \in M: \neg P(x)

Thus one can use the two quantifiers just as one would use \forall and \exists usually. Note however than \forall and \tilde{\forall} don't interchange and neither do \exists and \tilde{\exists}.

One last note: it might be worth using a different notation, for example: \exists ^ \infty might be easier to understand than \tilde{\exists} , and \forall ^ \approx might be better than \tilde{\forall}.

I'll try to formalize these quantifiers some more when I find some spare time.

Stay tuned for coding related updates soon :-)
Cheers,
Andreas

Extracting Information from StudiVZ

Posted by on October 19, 2009 1 comment

Some time ago somebody stole 1 million data records from StudiVZ, the German Facebook clone. I'm not exactly sure why people call the person a hacker who stole data, because it appears he simply wrote a tool that harvested the publicly available data from StudiVZ (which everyone with an account can view).

People on StudiVZ share all their data by default---contrary to Facebook which values a person's private data a lot more. Thus by simply opening each profile from a dummy user and processing the HTML data from StudiVZ one can extract a lot and some more information from random people who probably don't even know about it or don't care.. so I'm not sure about the stealing part.

Apparently there are some captcha's when you start browsing searches beyond a few pages. I guess that is where the hacking part comes in, because getting around a captcha probably constitutes hacking---maybe?

Anyway I think part of the media coverage is a bit ridiculous because anyone can write a simple harvester in an hour or two. It took me one and half hours, so I think I'm on the safe side with this estimate and I didn't really have a clue about this stuff before either.

Since I don't want to "hack", I've only written a very tame harvester. It connects to your personal StudiVZ account, and retrieves the name and profile ID (and thus profile URL) of all your friends in the "Meine Freunde" pages.

It could do a lot more with that like retrieving everybody's birthday or random pictures, but I'm too lazy to code that because you use the same pattern for extracting data over and over again and it stops being interesting quite fast.

You can download the project here. It is a one file C# project. I'm releasing it under GPL (whatever).

It's really easy to explain how it works:

  • It uses System.Net's HttpWebRequest and HttpWebResponse to get (and post) web pages.
  • StudiVZ (like every other portal) uses cookies, so I create a CookieContainer and use it in every http request.
  • There are a few hidden values that StudiVZ expects during login. I'm retrieving them from the main page using custom built regular expressions. I've found a handy AJAX tester for .NET regular expressions. It was really useful for building the expressions and debugging them. (BTW you can find all URLs I used in the comments.)
  • After login I use the same pattern: get page & parse using regex for everything.
  • Visual Studio has an awesome "HTML Visualizer" for strings. It displays the content of a string as HTML page, which is really nifty if you're doing anything related to HTML processing.

The code is quite ugly. Well, it's not production code and this is only meant as a proof of concept.

Also note that I have at most violated the AGB of StudiVZ and not committed any criminal acts and I'm not planning to sell my friend's profile IDs or data either :-)

Maybe someone can extend the code and make it more useful. I guess it would be fun to automatically download all your pictures (including tags) and feed them into flickr or picasa... but someone else can do that.

Cheers,
Andreas

Array