Wednesday, December 23, 2009

// // Leave a Comment

Сватбата - експериментална комедия (The Wedding, a short film)

Сватбата - експериментална комедия (The Wedding) from Todor Arnaudov on Vimeo.

Във Вимео е най-добро качество, но ако не работи:

Български - виж по-долу.

"The Wedding" - my first complex movie. (The film is in Bulgarian, not yet subtitles)
Genre: experimental surrealistic comedy.
Synopsis: a young reporter is visiting a mass wedding of the Sect of FMI to make a report, however he makes the terrifying discovery that he's also supposed to be married, without knowing who. The nightmare goes worst, when he finds out that everybody seems to be making fun of him. However, how it is going to end?

Млад репортер отива да заснеме масова сватба в Сектата на ФМИ в Пловдив, където прави ужасяващото откритие, че той също е заплануван за Сватбата и не се знае за коя ще го оженят. Това е само началото на неговия кошмар, защото изглежда, че всички се подиграват с него... Как обаче ще завърши всичко?

Повече инфо в блога на Twenkid Studio
Read More

Wednesday, December 16, 2009

// // Leave a Comment

Reinforcement learning - comments on a Ben Goertzel's blog article

About: "Reinforcement Learning: Some Limitations of the Paradigm"

Bulgarian readers could also check out my article: Анализ на смисъла на изречение въз основа на базата знания на действаща мислеща машина. Мисли за смисъла и изкуствената мисъл.


You're right that if the system can alter or "re-wire" its circuitry that computes "reward", this confuses the mechanism, but I believe humans actually can do that - person's values are changing during life time, and even during the decision processes themselves.

I think "reward" is looked in a too narrow sense, because mind is not as solid and... single-minded to have one-single type of rewards. You've spoken about the "sub-selves" in the blog, I would say "virtual control units" that take control over the body.

Body is what makes mind to look uniform, even if it's not, mind can want 1000 things, body can't do them in the same time.

I've speculated on that in my old article, but it's in Bulgarian, I have to translate it in English (eventyally citing it).

E.g. let's consider a boy which is hesitating whether to eat up a chocolate or not. This could be an immediate high taste reward in a near future and if the boy plans only for 1 minute ahead, this is a right decision.

However, what if the boy widen the period of prediction to one year? He remembers his pain while visiting his dentists and reminds his notes, that eating too much chocolate causes bad teeth and pain. So if he plans for one year and reminds this, and decides that this will happen (it couldn't be sure), then the highly rewarding decision would stop being rewarding in the equation and the boy wouldn't take it.

Overall, the maximum reward depends on the set of predicting sub-units, scenarios, values taken into account in the very moment of decision, and the period of time they are predicting ahead.

They are changing, depending on attention, context, mood or even chance - there are so many scenarios that the brain can think of.

The set of predicting sub-units may change, they can switch at different levels of hierarchies and predicted periods, while the behaviour could still keep being reward-driven. It could be reward-driven for the particular "reward-driven virtual unit that took control over the body in this very moment".

This implies the mind is not uniform and there is not single "greatest reward".


And another comment on another comment:

I mean, e.g. if you're 18, you may think that making random sex right now is "cool". When you are 28, you may think it's not, even though making sex still would bring you immediate pleasere - however, higher level controls would inhibit your urge for lower level rewards and generally shift your behaviour to higher level rewards.

My point is that reward is not only dopamine, endorphin or so. Higher the ingelligence, higher the abstraction of reward could be. A reward is what the one that receives it considers "a reward", and even the altruism could also be taken as egoism, because one is doing what is "good" regarding its own values and desires, sometimes it's against what the other wants.

E.g. when a lover dies to save his beloved one. Is it really an altruism? How she will feel when seeing him dying, wouldn't she prefer them to die together?

And if you're doing something "against yourself" isn't it to prevent something that you consider worst. When you feel moral responsibility of doing something, fear of not fulfilling your duty might be bigger than the fear of pain.


На български - учене с подкрепление, машинно обучение, изкуствен интелект, силно направление, Бен Гьорцел
Read More

Saturday, December 12, 2009

// // Leave a Comment

How to merge/append/join wav segments/files into one?

Appending two or a few Wav files is easy with sound editors such as Audacity, Cool Edit and SoundForge. However, what if you need to join hundreds of files in different directories? Doing this manually is insane...

Recently I needed to do this with a bunch of records in Italian I wanted to put in my mp3-player. I found only a commercial solution, but I decided to code mine.

Todor's WaveAppender 0.01

Platform:Windows.License: Freeware.Warranty: No.

Download:Wave Appender 0.01

Please, post a comment if you find it useful.


- Unzip files.
- Open scan.bat and edit the second string to point the path to the base folder with Wav files. E.g.:

scan.exe l:\italian\wave\1\ >list.txt

scan.exe your_full_path_here >list.txt

- Run scan.bat - it generates list.txt.
- Open list.txt and put on top a line pointing to the target directory. The merged files will be created there, e.g.:




- Run append.bat
- Voila! Files from each folder will be merged in 1.wav (first folder in the list), 2.wav, ... N.wav.

Using only appWav.exe

You can edit the list file manually, then you don't need to run the scanner.
The syntax is:

First line - target path
Next - N lists of path to files to be merged in N.wav. A list ends with "#".
The file ends with "$".







- Files in each merge group should have the same format, code is pretty simple yet.
- Files in one list can be from different folders.


Thanks to Srinivas Varukala (basic code for scanning directories in C#) and Vasian Cepa (Numeric Sorter in C#) for the Scanner part; you can find their code in CodeProject.

Appender code is in C++, written by me.

Keywords: How to merge wav files? How to append wav files. How to join wav files. Merging wav files.

Read More
// // Leave a Comment

Сборник с разкази и стихове с участието на Тош | A book with Todor's co-autorship

Last month was issued a compilation with prose and poetry, where Todor participates with two short stories.

"Три години БГлог", "3 години Бглог", "Авторът си ти"

Read More

Wednesday, December 9, 2009

// // Leave a Comment

Шофьорът на Москвич 2 - Тодор Арнаудов и Ангел Янев | The Moskvich Driver 2

Филм на Тош

Продължение на "Шофьорът на Москвич", който беше донякъде филмово начало на Twenkid Studio през 2008 г. и беше "ням", не го озвучих. Този път с Ангел, с говор, сочен звук и дори малко музика от мен.

"Шофьорът на Москвич 2" стана най-сетне и дебютът на Ангел. :) Има много хляб в него като комик.

Шофьорът на Москвич

Други ключови думи: москвич, 1500, АЗЛК, комедия, късометражни, филми, 408, 407, копач, джуган, таратайка, соц., комиците
Read More

Friday, December 4, 2009

// // Leave a Comment

DARPA PAL, CALO, RADAR - Cognitive Assistants that Learn and Organize

Other ambitious AGI projects, and these ones belong to DARPA. Scary!

I have a project for a similar tool that I call "Research Accelerator" or "Cognitive Accelerator", that initially would be an integration of various NLP, Machine Learning and Computer Vision techniques which assist and speed up in searching, reading, information extraction and any cognitive taks. This system also would be capable to work with GUI on its own, controlling the mouse and keyboard to do some job instead of user.

Time is ticking away and I need partners...

PAL - Personalized Assistant that Learns

RADAR - Reflected Agents with Distributed Adaptive Reasoning
CALO - Cognitive Assistant that Learns and Organizes
Read More