Adversarial Approach to Prediction in Atari Games

Marian Andrecki; Nicholas Kenelm Taylor

Adversarial Approach to Prediction in Atari Games

Marian Andrecki
, Nicholas Kenelm Taylor

Research output: Contribution to conference › Paper › peer-review

116 Downloads (Pure)

Abstract

Recent advances in neural networks have resulted in reliable predictions of observations in Atari games. The quality of predictions can be improved by addition of a generative adversarial approach to the training. This research investigates this development and the benefits of unsupervised predictive learning for reinforcement learning agents.

Original language	English
Number of pages	2
Publication status	Published - 7 Jun 2017
Event	2017 EPSRC CDT Student Conference – Oxford, Bristol and Edinburgh - Oxford, United Kingdom Duration: 7 Jun 2017 → 7 Jun 2017 http://aims.robots.ox.ac.uk/epsrc-cdt-student-conference-oxford-bristol-and-edinburgh/

Conference

Conference	2017 EPSRC CDT Student Conference – Oxford, Bristol and Edinburgh
Country/Territory	United Kingdom
City	Oxford
Period	7/06/17 → 7/06/17
Internet address	http://aims.robots.ox.ac.uk/epsrc-cdt-student-conference-oxford-bristol-and-edinburgh/

Access to Document

Download pdfFinal published version, 144 KBLicence: Unspecified

Cite this

@conference{09f6d1fd4cfb4722b056700874857451,

title = "Adversarial Approach to Prediction in Atari Games",

abstract = "Recent advances in neural networks have resulted in reliable predictions of observations in Atari games. The quality of predictions can be improved by addition of a generative adversarial approach to the training. This research investigates this development and the benefits of unsupervised predictive learning for reinforcement learning agents.",

author = "Marian Andrecki and Taylor, \{Nicholas Kenelm\}",

year = "2017",

month = jun,

day = "7",

language = "English",

note = "2017 EPSRC CDT Student Conference – Oxford, Bristol and Edinburgh ; Conference date: 07-06-2017 Through 07-06-2017",

url = "http://aims.robots.ox.ac.uk/epsrc-cdt-student-conference-oxford-bristol-and-edinburgh/",

}

TY - CONF

T1 - Adversarial Approach to Prediction in Atari Games

AU - Andrecki, Marian

AU - Taylor, Nicholas Kenelm

PY - 2017/6/7

Y1 - 2017/6/7

N2 - Recent advances in neural networks have resulted in reliable predictions of observations in Atari games. The quality of predictions can be improved by addition of a generative adversarial approach to the training. This research investigates this development and the benefits of unsupervised predictive learning for reinforcement learning agents.

AB - Recent advances in neural networks have resulted in reliable predictions of observations in Atari games. The quality of predictions can be improved by addition of a generative adversarial approach to the training. This research investigates this development and the benefits of unsupervised predictive learning for reinforcement learning agents.

M3 - Paper

T2 - 2017 EPSRC CDT Student Conference – Oxford, Bristol and Edinburgh

Y2 - 7 June 2017 through 7 June 2017

ER -

Adversarial Approach to Prediction in Atari Games

Abstract

Conference

Access to Document

Fingerprint

Cite this