Deep reinforcement learning for Flappy Bird using TensorFlowJS

dc.contributor.author	Kovtun, Matvii
dc.date.accessioned	2024-02-19T10:37:19Z
dc.date.available	2024-02-19T10:37:19Z
dc.date.issued	2019
dc.identifier.citation	Kovtun, Matvii. Deep reinforcement learning for Flappy Bird using TensorFlowJS / Kovtun, Matvii; Supervisor: Supervisor: Mykhailo Ivankiv; Ukrainian Catholic University, Department of Computer Sciences. – Lviv: 2019. – 26 p.	uk
dc.identifier.uri	https://er.ucu.edu.ua/handle/1/4560
dc.language.iso	en	uk
dc.title	Deep reinforcement learning for Flappy Bird using TensorFlowJS	uk
dc.type	Preprint	uk
dc.status	Публікується вперше	uk
dc.description.abstracten	In this paper, I will cover a specific topic of Deep Reinforcement Learning which will be performed in the browser. I will provide an architectural overview of a system, describe a process of learning, show a deployment process. There are two key parts in this work - environment of an execution which is a browser and a learning approach which is Deep Reinforcement Learning. Motivation for this was a rapid development in both of these technologies deep learning and web. Demonstration of my work can be found here: AI playing Flappy Bird Code can be found here: Github repository	uk