Evolving effective micro behaviors in RTS game

Liu, Siming; Louis, Sushil J.; Ballinger, Christopher

doi:10.1109/cig.2014.6932904

Cited by 10 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the context of StarCraft micromanagement, a large spectrum of AI approaches have been studied. There has been work on Bayesian fusion of hand-designed influence maps [36], fast heuristic search (in a simplified simulator of battles without collisions) [5], and even evolutionary optimization [16]. Closer to this work, [45] successfully applied tabular Q-learning [44] and SARSA [34], with and without experience replay ("eligility traces"), with a reward similar to the one used in several of our experiments.…”

Section: Related Workmentioning

confidence: 94%

Episodic Exploration for Deep Deterministic Policies: An Application to StarCraft Micromanagement Tasks

Usunier¹,

Synnaeve²,

Lin³

et al. 2016

Preprint

View full text Add to dashboard Cite

We consider scenarios from the real-time strategy game StarCraft as new benchmarks for reinforcement learning algorithms. We propose micromanagement tasks, which present the problem of the short-term, low-level control of army members during a battle. From a reinforcement learning point of view, these scenarios are challenging because the stateaction space is very large, and because there is no obvious feature representation for the state-action evaluation function. We describe our approach to tackle the micromanagement scenarios with deep neural network controllers from raw state features given by the game engine. In addition, we present a heuristic reinforcement learning algorithm which combines direct exploration in the policy space and backpropagation. This algorithm allows for the collection of traces for learning using deterministic policies, which appears much more efficient than, for example, -greedy exploration. Experiments show that with this algorithm, we successfully learn non-trivial strategies for scenarios with armies of up to 15 agents, where both Q-learning and REINFORCE struggle.

show abstract

Section: Related Workmentioning

confidence: 94%

Episodic Exploration for Deep Deterministic Policies: An Application to StarCraft Micromanagement Tasks

Usunier¹,

Synnaeve²,

Lin³

et al. 2016

Preprint

View full text Add to dashboard Cite

show abstract

“…There are also multiple studies focused on player modeling. Siming Liu et al [16] recognized players through extracted features and Random Forest. T. Avontuur et al [6] developed a model that predicts skills based on data collected during the early portions of the game.…”

Section: Related Workmentioning

confidence: 99%

Feature Extraction for StarCraft II League Prediction

Lee

Ahn

2021

Electronics

View full text Add to dashboard Cite

In a player-versus-player game such as StarCraft II, it is important to match players with others with similar skills. Studies modeling player skills were conducted, with 47.3% and 61.3% performance. In order to improve the performance, we collected 46,398 replays and compared features extracted from six sections of replays. Through the comparison of the six datasets we created, we propose a method for extracting features from a single replay. Two algorithms, k-Nearest Neighbors and Random Forest, which are most commonly used in related studies, are compared. Our research showed a outperforming accuracy of 75.3% compared to previous works. Although no direct comparison has been made with the current system, we conclude that our research can replace the placement games of five rounds.

show abstract

“…They show that it is an essential part of tactical behavior to select the right enemy to attack. The work of [Liu et al 2014] presents an automatic target selection approach using Genetic Algorithms. However, their concept of Target Selection, although still tactic, shows a different concept from [Uriarte and Ontañón 2012].…”

Section: Target Selectionmentioning

confidence: 99%

On the Development of Intelligent Agents for MOBA Games

Silva

Chaimowicz

2015

2015 14th Brazilian Symposium on Computer Games and Digital Entertainment (SBGames)

View full text Add to dashboard Cite

Multiplayer Online Battle Arena (MOBA) is one of the most played game genres nowadays. With the increasing growth of this genre, it becomes necessary to develop effective intelligent agents to play alongside or against human players. In this paper we address the problem of agent development for MOBA games. We implement a two-layered architecture agent that handles both navigation and game mechanics. This architecture relies on the use of Influence Maps, a widely used approach for tactical analysis. Several experiments were performed using League of Legends as a testbed, and show promising results in this highly dynamic real-time context.

show abstract

Evolving effective micro behaviors in RTS game

Cited by 10 publications

References 15 publications

Episodic Exploration for Deep Deterministic Policies: An Application to StarCraft Micromanagement Tasks

Episodic Exploration for Deep Deterministic Policies: An Application to StarCraft Micromanagement Tasks

Feature Extraction for StarCraft II League Prediction

On the Development of Intelligent Agents for MOBA Games

Contact Info

Product

Resources

About