Learning bird (Demo)

A naive bird that learns how to flap by NeuroEvolution.

NeuroEvolution

Neuroevolution, or neuro-evolution, is a form of artificial intelligence that uses evolutionary algorithms to generate artificial neural networks (ANN), parameters, topology and rules. More info.

Algorithms

Neural Network

• A fully-connected feedforward neural network with one hidden layer [5 4 1].

• Inputs:

  • y position of a bird.

  • y position of nearest top pipe.

  • y position of nearest bottom pipe.

  • x position of neaset pipe.

  • Velocity of a bird.

• Output:

  • output > 0.5 then bird.flap().

Genetic Algorithm

Step 1: Create a population.

Step 2: LOOP

• Evaluate the fitness of birds' brains (aka neural networks)

• Create a new generation.

• Pick parents based on their fitness score to breed.

  • Crossover

  • Mutation

Here are parameters used in the demo:

• Population: 100
• Elitism: 0.2 (best birds are kept unchanged for the next generation)
• New random birds: 0.2
• Mutation rate: 0.2
• Mutation range: 0.5