A Minimal Training Strategy to Play Flappy Bird Indefinitely with NEAT
- 1 October 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 21-28
Abstract
A large number of algorithms to generate behaviors of game agents have been developed in recent years. Most of them are based on artificial intelligence techniques that need a training stage. In this context, this paper proposes a minimal training strategy to develop autonomous virtual players using the NEAT neuroevolutionary algorithm to evolve an agent capable of playing the Flappy Bird game. NEAT was used to find the simplest neural network architecture that can perfectly play the game. The modeling of the scenarios and the fitness function were set to ensure adequate representation of the problem compared to the real game. The fitness function is a weighted average based on multiple scenarios and scenario-specific components. Coupling the minimal training strategy, a representative fitness and NEAT, the algorithm had a short convergence time (around 20 generations), with a low complexity network and achieved the perfect behavior in the game.Keywords
This publication has 10 references indexed in Scilit:
- Browser/Server based Experimental Environment for Reinforcement Learning2018
- Type-2 fuzzified flappy bird control system2016
- Deduction of fighting game countermeasures using Neuroevolution of Augmenting Topologies2016
- Python code generation for explicit MPC in MPT2016
- Neuroevolution for General Video Game Playing2015
- Shaping in reinforcement learning via knowledge transferred from human-demonstrations2015
- Discovering Multimodal Behavior in Ms. Pac-Man Through Evolution of Modular Neural NetworksIEEE Transactions on Computational Intelligence and AI in Games, 2015
- Automatic Content Generation in the Galactic Arms Race Video GameIEEE Transactions on Computational Intelligence and AI in Games, 2009
- Real-Time Neuroevolution in the NERO Video GameIEEE Transactions on Evolutionary Computation, 2005
- Evolving a Roving Eye for Go2004