## Text to picture

Hey guys, I created a web visualisation for w2vec translated text. It basically generates a picture from your text using w2vec representations for words and a chosen palette. [Continue Reading..]

## The road to wisdom, bfgs nlp

First, let me tell you about the newsletter button I added :D . You can subscribe to get notified when I post :). It’s down and right, please feel compelled to press it :D .

The road to wisdom?—Well, it’s plain

and simple to express:

Err

and err

and err again,

but less

and less

and less.

— Piet Hein

The above words are from a grook I like.

Sometimes I spend all day trying to count

the leaves on a single tree. To do this I

have to climb branch by branch and

write down the numbers in a little book.

So I suppose, from their point of view,

it’s reasonable that my friends say: what

foolishness! She’s got her head in the clouds

again.

But it’s not. Of course I have to give up,

but by then I’m half crazy with the wonder

of it — the abundance of the leaves, the

quietness of the branches, the hopelessness

of my effort. And I am in that delicious

and important place, roaring with laughter,

full of earth-praise.

-Mary Oliver,Foolishness? No, it’s not

Autumn is great in Romania, lots of beautiful light and nice colored leafs. I enjoyed great walks in the park, but naturally I started thinking about: what would it take for a machine to detect leafs in a picture. I actually don’t know how natural this thought is :)) but I said it’s a good problem for the weekend, brushing up on my maths and maybe if I could write it in a tutorial manner it could help somebody.

This article will not be about poems, of course, but instead it will be about implementing a technique that enables you to get to where you want to be :). Following the MUI article I felt like I needed to do something about nonlinear optimisations. [Continue Reading..]

MIUIII

# Play the puzzle!

The Goal: Given a starting “sentence” (string) reach the goal “sentence” (string) by only applying the rules below.

Rule Example
Mx -> Mxx: MIIU -> MIIUIIU
xI -> xIU: MIIUI -> MIIUIU
xIIIy -> xy: MUIUIIIU -> MUIUUU
xUUy -> xy: MUIUUIIIU -> MUIIIIU
• x and y represent any character strings and M, U, I represent the actual letters, the arrow may be read as “becomes”.

• For example the first rule can be read: Any string that ends in M becomes M and two times the reminder.

Try it, I will be waiting here :). [Continue Reading..]

## Evolving the best driver using differential evolution

Children (light blue) and parents(dark blue) run in the same generation simulation. The red car is the current best individual, the white ones are the old best individual (evaluated in the same generation). [Continue Reading..]

## On rational parametrization of circles

In programming and engineering in general we need to generate circles or circle like shapes. Often we need the circle points for all kinds of stuff, like drawing or object placement or procedural generation, interpolation etc. Let’s take a ... [Continue Reading..]

## Implementing a simple top down racing game ai controller

We have the procedural track, we have the ghysics car model, let’s make a simple top down racing AI! Above you can see me play with the basic AI (I am the red one that always ends up hitting the wall :) ) [Continue Reading..]

## Top down racing car physics model using box2D

Today’s article is about implementing a simple, easy configurable car model with the Box2D physics library. See above video of me playing with the new model. [Continue Reading..]

Next → Page 1 of 2

· This is not your average Footer