Self-learning game bot using deep reinforcement learning
This thesis report is submitted in partial fulfilment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2017.
| 1. Verfasser: | |
|---|---|
| Weitere Verfasser: | |
| Format: | Abschlussarbeit |
| Sprache: | English |
| Veröffentlicht: |
BRAC University
2018
|
| Schlagworte: | |
| Online Zugang: | http://hdl.handle.net/10361/9509 |
| id |
10361-9509 |
|---|---|
| record_format |
dspace |
| spelling |
10361-95092022-01-26T10:10:24Z Self-learning game bot using deep reinforcement learning Ananto, Azizul Haque Mostakim, Moin Department of Computer Science and Engineering, BRAC University Game Bot Self-learning Reinforcement learning This thesis report is submitted in partial fulfilment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2017. Cataloged from PDF version of thesis report. Includes bibliographical references (pages 45-47). We present a deep learning model for playing games with high level input (image/raw pixel) using reinforcement learning. The games are action limited (like snakes, catcher, air-raider etc.). The model consists of convolution neural network for processing image inputs and fully connected layers for estimating actions according to the inputs where the idea of taking action is based on Q-learning (model-free reinforcement learning), yet modified it for our policy and usage. We applied our method on the python’s ‘PyGame Learning Environment’ and some other classic control games. We found our method learns fast enough but not with best accuracy. Then we tried the batch of input method which results a high score for the Catcher environment. It produced better performance in terms of the learning speed and accuracy. Azizul Haque Ananto B. Computer Science and Engineering 2018-02-19T05:29:34Z 2018-02-19T05:29:34Z 2017 2017-12 Thesis ID 14301050 http://hdl.handle.net/10361/9509 en BRAC University thesis reports are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. 47 pages application/pdf BRAC University |
| institution |
Brac University |
| collection |
Institutional Repository |
| language |
English |
| topic |
Game Bot Self-learning Reinforcement learning |
| spellingShingle |
Game Bot Self-learning Reinforcement learning Ananto, Azizul Haque Self-learning game bot using deep reinforcement learning |
| description |
This thesis report is submitted in partial fulfilment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2017. |
| author2 |
Mostakim, Moin |
| author_facet |
Mostakim, Moin Ananto, Azizul Haque |
| format |
Thesis |
| author |
Ananto, Azizul Haque |
| author_sort |
Ananto, Azizul Haque |
| title |
Self-learning game bot using deep reinforcement learning |
| title_short |
Self-learning game bot using deep reinforcement learning |
| title_full |
Self-learning game bot using deep reinforcement learning |
| title_fullStr |
Self-learning game bot using deep reinforcement learning |
| title_full_unstemmed |
Self-learning game bot using deep reinforcement learning |
| title_sort |
self-learning game bot using deep reinforcement learning |
| publisher |
BRAC University |
| publishDate |
2018 |
| url |
http://hdl.handle.net/10361/9509 |
| work_keys_str_mv |
AT anantoazizulhaque selflearninggamebotusingdeepreinforcementlearning |
| _version_ |
1814307707898822656 |