Half Moon Project Part 2 - More Accuracy

-
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.



Comments

Post a Comment

Popular updates

Braitenberg Vehicles – v1.0

-
Image
For our Braitenberg Vehicles project, we were to create Braitenberg's first four cars. The basic idea behind these vehicles is that each vehicle has a set of sensors and motors. The sensors cause the motors to move: either speed up or slow down. By connecting different sensors to different motors, you can cause the cars to have different movements. Math: Math!! Although I usually start with the visual side of these projects, this time I wanted to first figure out how the vehicles would move on-screen. Getting Around The first car, Getting Around , was pretty simple. The light intensity, taken from the sensor, directly correlates to the y-position of the car. At this point, I threw together some basic graphics using the handy-dandy StdDraw library and a few JPEGs. You can see the final Getting Around  car to the right. Implementing two motors and sensors, however, was much more challenging. At first, I was trying to calculate the distance each motor would travel ...
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 adde...
Read more