High-Level Strategy Selection under Partial Observability in StarCraft: Brood War

11/21/2018
by   Jonas Gehring, et al.
2

We consider the problem of high-level strategy selection in the adversarial setting of real-time strategy games from a reinforcement learning perspective, where taking an action corresponds to switching to the respective strategy. Here, a good strategy successfully counters the opponent's current and possible future strategies which can only be estimated using partial observations. We investigate whether we can utilize the full game state information during training time (in the form of an auxiliary prediction task) to increase performance. Experiments carried out within a StarCraft: Brood War bot against strong community bots show substantial win rate improvements over a fixed-strategy baseline and encouraging results when learning with the auxiliary task.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
11/30/2018

Forward Modeling for Partial Observation Strategy Games - A StarCraft Defogger

We formulate the problem of defogging as state estimation and future sta...
research
03/18/2021

Canonical Representations for Direct Generation of Strategies in High-level Petri Games (Full Version)

Petri games are a multi-player game model for the synthesis problem in d...
research
02/28/2020

Reinforcement Learning in FlipIt

Reinforcement learning has shown much success in games such as chess, ba...
research
10/16/2012

Inferring Strategies from Limited Reconnaissance in Real-time Strategy Games

In typical real-time strategy (RTS) games, enemy units are visible only ...
research
09/11/2017

Combining Strategic Learning and Tactical Search in Real-Time Strategy Games

A commonly used technique for managing AI complexity in real-time strate...
research
07/31/2021

Inverse Reinforcement Learning for Strategy Identification

In adversarial environments, one side could gain an advantage by identif...
research
12/15/2021

Mask-combine Decoding and Classification Approach for Punctuation Prediction with real-time Inference Constraints

In this work, we unify several existing decoding strategies for punctuat...

Please sign up or login with your details

Forgot password? Click here to reset