AI Conference and New Year's Resolutions..

     Being busy with placement year does not me with enough time and motivation for anything else. This is exactly why I decided to start tracking my progress and work I do in order to acquire understanding of AI concepts. Therefore I started marking down days I've done something AI related that includes things like reading AI research papers or completing courses materials. Each week I am also trying to track progress in comparison to the previous week, the ultimate goal is to keep progress chart green with gradually increasing AI content per week.

     I found that this as well as attending AI Conference in Cambridge organised by BCS Chapter increased my motivation a lot and now I feel I am using the time better.

     Attending AI Conference was amazing experience and I hope next year I will be able to go for the full three days of the conference rather than one day only (I'd better start saving now!). Meeting people working in AI industry, whether they were working for a company or at the university, was very motivational and inspirational.

     So as to my plan for 2016 it basically is:

• complete as many AI related courses as I can (this includes finishing the ones I've started first);
• start a new small project related to Neural Networks;
• continue with one of my second year projects I've done this year with my friend and hopefully end up having a game out of it with AI system;
• shape an idea for final year project by first reasearching the possibility of it and maybe consulting this with lecturers before starting final year;
• read more interesting AI related research papers and keep up to date with AI news;

Some more about my last project...

     Last weeks were quite hectic for me... Therefore I haven't really had time to progress with any of the Coursera courses at the moment. I will try to get back to those tomorrow, so you can expect some posts coming up soon.

     For now I will write about details of my first AI project. From the previous post you know that I had to train a Neural Network in Neuroph to control a car in TORCS application.

    A little bit more information about TORCS:
           TORCS is open application used for championships every year. It stands for The Open Racing Car Simulator. Each year people compete against each other to come up with the best controller they can. This is also used for research purposes (mainly by PhD students but it's not limited to them).
            To control a car via code TORCS has a number of variables used- those are described in TORCS official documentation. For my project I mainly used steering values as well as sensor values that has given distance of the car from the track.


     A bit about Neuroph:
               Neuroph is an open source framework that runs on top of NetBeans and is coded with Java. It is usefult for creating Neural Networks via GUI system and then testing them (depending what algorithms are used training data may be required as well). Then saved Neural network can be used again through Java code if using correct libraries.


More coming soon! :)

Machine Learning course from Stanford

     Today I have finished obligatory quizzes for Machine Learning course from Stanford I have started recently. Some questions in the second quiz were more tricky than I expected. But hopefully I will be able to follow up with the rest of the course.

     So far the teaching materials- video lectures were of an amazing quality. Everything is explained very well and even a beginner in the field of AI will be able to understand it. Appropriate examples that were given are also really helpful. So far it didn't even require that much maths knowledge but it definitely has given great overview of how some mathematical equations are used for machine learning algorithms (gradient descent algorithm specifically).

     I also really like the third part non-obligatory- an introduction to some operations on matrices, vectors etc. That's the last thing left for me to do before entering week 2 when the materials will be released.

PS: While waiting for week 2 materials to be released I will give some description of my last university AI project I've made. You might have seen an introduction about it in earlier posts.

Machine Learning course

   Since I still have couple days of holidays I plan to use them to learn about AI, to be more specific-Machine Learning. Therefore I have started taking part in Machine learning course on Coursera from Stanford University.

I have enrolled early and therefore had an access to the first week of the course already-I am now halfway through completing it. First quiz was quite easy and just checked the basic knowledge about supervised and unsupervised learning algorithms.

All in all the course seems to be well-organised and I am looking forward to the next week materials! :) So far I would definitely recommend it. More updated to come..

My very first AI based project..

      A while ago I had to carry out simple research for my AI module. At least it seemed simple at the beginning...

      The project initial idea was suggested by module leader tutor and was to train a Neural Network in Neuroph 2.9 to control a car in TORCS application (car racing application used for yearly AI competitions and Master/PhD level research mostly). Even though module's material covered neural network's concepts inspirations etc. it was not enough to properly understand the background calculations neural network was doing. Therefore this project required me to carry out a lot of background research about it.

      The project concluded with a written 'research paper' on carried out experiments. This was also an inspiration to begin Computational Neuroscience course on Coursera.

Welcome :)

    Welcome to the blog about Artificial Intelligence and anything connected to this topic. The journey into learning interesting but difficult concepts and their implementations. I may also be linking interesting AI articles and courses I would recommend or not so follow me for updates.