Difference between revisions of "AdjutantBot"

From StarCraft AI, the resource for custom StarCraft Brood War AIs
Jump to: navigation, search
 
Line 13: Line 13:
 
|externalDownload=
 
|externalDownload=
 
|language=C++
 
|language=C++
|googleCode=https://code.google.com/p/adjutantbot/
+
|googleCode=
 +
|github=https://github.com/albertouri/adjutantbot/
 
|site=
 
|site=
 
|tlstream=
 
|tlstream=
Line 29: Line 30:
 
|achievements=
 
|achievements=
 
}}
 
}}
 +
 +
This project focuses on the problem of avoiding troop loss by identifying good tactical groupings. Banding separated units together using UCT (Upper Confidence bounds applied to Trees) along with a learned reward model outperforms grouping heuristics at winning battles while preserving resources. Our Adjutant bot design won the best Newcomer honor at CIG 2012
 +
 +
==Videos==
 +
{{#ev:youtube|PfeB--1qaww}}
 +
 +
==Scientific Publications==
 +
* [http://ial.eecs.ucf.edu/Sukthankar-CIG2013.pdf Adjutant Bot: An Evaluation of Unit Micromanagement Tactics]. Nicholas Bowen*, Jonathan Todd*, and Gita Sukthankar. IEEE Conference on Computational Intelligence in Games (Competition). 2013.

Latest revision as of 15:17, 23 July 2015

[e][h]Ticon.png AdjutantBot
Author(s):
Nicholas Bowen
Affiliation:
University of Central Florida
Country:
USA USA
Race:
ELO peak:
BWAPI version:
3.7.4
Type:
DLL
Download:
Language:
C++
Source code:
Github.png

This project focuses on the problem of avoiding troop loss by identifying good tactical groupings. Banding separated units together using UCT (Upper Confidence bounds applied to Trees) along with a learned reward model outperforms grouping heuristics at winning battles while preserving resources. Our Adjutant bot design won the best Newcomer honor at CIG 2012

Videos

Scientific Publications