Genetic Machine Learning

-
Genetic Machine Learning is another type of ML, completely separate from Neural Networks.

Evolution-Based

Evolution is complex. Learning to walk, for example, takes millions upon millions of years of small mutations, and survival of the fittest. If an animal mutates, gains an advantage, and lives longer than others, then it has more kids. Those kids will probably also have that advantage.

That is the way genetics-based ML works. It starts with a randomly generated population and calculates the level of Fitness for each individual. The fittest of the bunch move on and have "kids", basically genetically-combined versions of themselves.

Landing a Rocket

The first part of this project was to create a rocket game - with a gravitational force, the force of the fuel, and graphics (?).

The rocket would fall at a certain speed (Gravity) and would be offset by the force of the fuel being used (a level between 0-8).

The Genetic Machine Learning algorithm would then figure out when to switch the level of fuel, and what level of fuel to use. It would start off random, but then the best of the lot would move on.

The fitness function in my program took into account whether the rocket crashed, whether it landed, the distance from the ground, the velocity at the end of the game, and the time it took to end the game. The game would also end with a penalty if the rocket went too high (off the screen).

After determining the best two rocket games, they would take the "genes" of the game (what levels the fuel was set to and when) and randomly combine them to make kids. There would also be a 1% chance of mutation, where the gene would be randomly generated again.

With this algorithm, the best versions of it were able to start seeing rockets land after the sixth or seventh generation!

Comments

Post a Comment

Popular updates

Half Moon Project Part 2 - More Accuracy

-
Image
This time around, we used the Neuroph library to create a neural network  that can divide the double half-moons more accurately. Neural Networks A Neural Network is a combination of Neurons that feed into each other. Using some really complex math, they can weight each input (from a previous Neuron) differently. Then, after training, they can use backpropagation to calculate which weights should be edited. What this looks like can be seen below. More Accuracy for Half-Moons Since, previously, we were only using one neuron to calculate the line to separate the two moons, we could only draw a linear line. This time, with a more complex neural network, it can separate the two with a function that's super complicated. Theoretically, a human could also derive the equation, but it's so complex that at this point, it's easier to get the Neural Network to do it for us.
Read more

NumberVision – Machine Learning Application

-
Image
The next way we used our Neuron class was to classify digits. Using a UI, our program was assigned to take an input digit (a 5x7 grid, each box has a shade of grey between black and white) and classify the digit as a number between 0 and 9. A Smarter Neuron Graphical depiction of a Neuron receiving more than 2 inputs. Previously, I had only been using a Neuron with two inputs: an X and Y coordinate. Previously, this allowed me to plot my points on a plane like I did with the Half Moon project. However, due to the 5x7 grid of inputs I was getting, I had to alter my Neuron class to be able to take in 35 inputs. Mostly, this involved changing a few numbers.  Applying the Neuron to NumberVision The digit '1', with progressively more noise added from left to right. Good algorithms should be able to generate a result even if the input is a little distorted. I applied the same goal to NumberVision. The flow chart above shows various levels of added distortio
Read more

Machine Learning & Half Moon Project

-
Image
This was our first Machine Learning assignment. Machine Learning is something that I was (and still am) super excited about. Basically, the algorithm 'learns' something, and can then apply it in other situations. Learning & Neurons This was the toughest concept to learn (ha). We started with a Neuron, similar to the neurons in our brains. The neuron can trigger (yes/no) based on a series of inputs, in our case two inputs. The equation for whether or not the neuron triggers is SUMMATION(Weight(n) * Input(n)) > threshold . Each input is weighted differently, and the weights are what we can change to 'train' the neuron. We use this equation:  ΔWeight = learningSpeed * (target - output) * input . For example, if we start with a speed of 0.2, an input of 1, a weight of 0.5, and a target of 0, the weight would change by -0.2 (0.2 * (1 - 0) * 1). Now, with a weight of 0.3, repeat the process. Once each input returns the desired output, the best weights have been fo
Read more