Scaling Tangled Program Graphs to Visual Reinforcement Learning in ViZDoom

Created by W.Langdon from gp-bibliography.bib Revision:1.8051

@InProceedings{Smith:2018:EuroGP,

• author = "Robert J. Smith and Malcolm I. Heywood",
• title = "Scaling Tangled Program Graphs to Visual Reinforcement Learning in {ViZDoom}",
• booktitle = "EuroGP 2018: Proceedings of the 21st European Conference on Genetic Programming",
• year = "2018",
• month = "4-6 " # apr,
• editor = "Mauro Castelli and Lukas Sekanina and Mengjie Zhang and Stefano Cagnoni and Pablo Garcia-Sanchez",
• series = "LNCS",
• volume = "10781",
• publisher = "Springer Verlag",
• address = "Parma, Italy",
• pages = "135--150",
• organisation = "EvoStar, Species",
• keywords = "genetic algorithms, genetic programming",
• isbn13 = "978-3-319-77552-4",
• DOI = "doi:10.1007/978-3-319-77553-1_9",
• abstract = "A tangled program graph framework (TPG) was recently proposed as an emergent process for decomposing tasks and simultaneously composing solutions by organizing code into graphs of teams of programs. The initial evaluation assessed the ability of TPG to discover agents capable of playing Atari game titles under the Arcade Learning Environment. This is an example of visual reinforcement learning, i.e. agents are evolved directly from the frame buffer without recourse to hand designed features. TPG was able to evolve solutions competitive with state-of-the-art deep reinforcement learning solutions, but at a fraction of the complexity. One simplification assumed was that the visual input could be down sampled from a 210 x 160 resolution to 42 x 32. In this work, we consider the challenging 3D first person shooter environment of ViZDoom and require that agents be evolved at the original visual resolution of 320 x 240 pixels. To do so, we address issues with task scenarios performing fitness evaluation over multiple tasks. The resulting TPG solutions retain all the emergent properties of the original work as well as the computational efficiency. Moreover, solutions appear to generalize across multiple task scenarios, whereas equivalent solutions from deep reinforcement learning have focused on single task scenarios alone.",
• notes = "Part of \cite{Castelli:2018:GP} EuroGP'2018 held in conjunction with EvoCOP2018, EvoMusArt2018 and EvoApplications2018",

}

Genetic Programming entries for Robert J Smith Malcolm Heywood

Citations