Abdessamed Ouessai | Mascara Mustafa Stambouli (original) (raw)

Uploads

Papers by Abdessamed Ouessai

Entertainment Computing, 2022

Real-Time Strategy (RTS) games are well-known for their substantially large combinatorial decisio... more Real-Time Strategy (RTS) games are well-known for their substantially large combinatorial decision and state spaces, responsible for creating significant challenges for search and machine learning techniques. Exploiting domain knowledge to assist in navigating the expansive decision and state spaces could facilitate the emergence of competitive RTS game-playing agents. Usually, domain knowledge can take the form of expert traces or expert-authored scripts. A script encodes a strategy conceived by a human expert and can be used to steer a search algorithm, such as Monte Carlo Tree Search (MCTS), towards high-value states. However, a script is coarse by nature, meaning that it could be subject to exploitation and poor low-level tactical performance. We propose to perceive scripts as a collection of heuristics that can be parameterized and combined to form a wide array of strategies. The parameterized heuristics mold and filter the decision space in favor of a strategy expressed in terms of parameters. The proposed agent, ParaMCTS, implements several common heuristics and uses NaïveMCTS to search the downsized decision space; however, it requires a preceding manual parameterization step. A genetic algorithm is proposed for use in an optimization phase that aims to replace manual tuning and find an optimal set of parameters for use by EvoPMCTS, the evolutionary counterpart of ParaMCTS. Experimentation results using the µRTS testbed show that EvoPMCTS outperforms several state-of-the-art agents across multiple maps of distinct layouts.

VI Congress of the Spanish Society for Video Game Sciences, 2020

The core challenge facing search techniques when used to play Real-Time Strategy (RTS) games is t... more The core challenge facing search techniques when used to play Real-Time Strategy (RTS) games is the extensive combinatorial decision space. Several approaches were proposed to alleviate this dimensionality burden, using scripts or action probability distributions, based on expert knowledge. We propose to replace expert-authored scripts with a collection of smaller parametric scripts we call heuristics and use them to pre-select actions for Monte Carlo Tree Search (MCTS). The advantages of this proposal consist of granular control of the decision space and the ability to adapt the agent's strategy in-game, all by altering the heuristics and their parameters. Experimentation results in µRTS using a proposed implementation have shown a significant performance gain over state-of-the-art agents.

IEEE CoG, 2020

The impressive performance of Monte Carlo Tree Search (MCTS) based game-playing agents in high br... more The impressive performance of Monte Carlo Tree Search (MCTS) based game-playing agents in high branching-factor domains such as Go, motivated researchers to apply and adapt MCTS to even more challenging domains. Real-time strategy (RTS) games feature a large combinatorial branching factor and a real-time aspect that pose significant challenges to a broad spectrum of AI techniques, including MCTS. Various MCTS enhancements were proposed, such as the combinatorial multi-armed bandit (CMAB) based sampling, state/action abstractions, and machine learning. In this paper, we propose to employ move pruning as a way to improve the performance of MCTS-based agents in the context of RTS games. We describe a class of possibly detrimental player-actions and propose several pruning approaches targeting it. The experimentation results in µRTS indicate that this could be a promising direction.

ICTAACS'19, 2019

Online planning is an important research area focusing on the problem of real-time decision makin... more Online planning is an important research area focusing on the problem of real-time decision making, using information extracted from the environment. The aim is to compute, at each decision point, the best decision possible that contributes to the realization of a fixed objective. Relevant application domains include robotics, control engineering and computer games. Real-time strategy (RTS) games pose considerable challenges to artificial intelligence techniques, due to their dynamic, complex and adversarial aspects, where online planning plays a prominent role. They also constitute an ideal research platform and test-bed for online planning. µRTS is an open-source AI research platform that features a minimalistic, yet complete RTS implementation, used by AI researchers for developing and testing intelligent RTS game-playing agents. The unique characteristics of µRTS helped for the emergence of interesting online adversarial planning techniques, dealing with multiple levels of abstraction. This paper presents the major µRTS online planning approaches to date, categorized by the degree of abstraction, in fully and partially observable environments.

ISPS'15, 2015

Web page classification based on topic or sentiments is a common application of web content minin... more Web page classification based on topic or sentiments is a common application of web content mining techniques. In this paper we will present a novel application based on targeting a nation. The aim is to be able to automatically distinguish websites targeting a specific nation, using both the URL and the content of a web page. In this paper we will address the issue of identifying Algerian-interest web pages using a machine learning approach.We will present the process of acquiring data for the supervised learning phase and adapting it into a usable dataset, as well as using it to construct three distinct classifiers using different parts of the data. The resulting classifiers have shown outstanding performances (up to F-score = 0.93) for such application.

Entertainment Computing, 2022

Real-Time Strategy (RTS) games are well-known for their substantially large combinatorial decisio... more Real-Time Strategy (RTS) games are well-known for their substantially large combinatorial decision and state spaces, responsible for creating significant challenges for search and machine learning techniques. Exploiting domain knowledge to assist in navigating the expansive decision and state spaces could facilitate the emergence of competitive RTS game-playing agents. Usually, domain knowledge can take the form of expert traces or expert-authored scripts. A script encodes a strategy conceived by a human expert and can be used to steer a search algorithm, such as Monte Carlo Tree Search (MCTS), towards high-value states. However, a script is coarse by nature, meaning that it could be subject to exploitation and poor low-level tactical performance. We propose to perceive scripts as a collection of heuristics that can be parameterized and combined to form a wide array of strategies. The parameterized heuristics mold and filter the decision space in favor of a strategy expressed in terms of parameters. The proposed agent, ParaMCTS, implements several common heuristics and uses NaïveMCTS to search the downsized decision space; however, it requires a preceding manual parameterization step. A genetic algorithm is proposed for use in an optimization phase that aims to replace manual tuning and find an optimal set of parameters for use by EvoPMCTS, the evolutionary counterpart of ParaMCTS. Experimentation results using the µRTS testbed show that EvoPMCTS outperforms several state-of-the-art agents across multiple maps of distinct layouts.

VI Congress of the Spanish Society for Video Game Sciences, 2020

The core challenge facing search techniques when used to play Real-Time Strategy (RTS) games is t... more The core challenge facing search techniques when used to play Real-Time Strategy (RTS) games is the extensive combinatorial decision space. Several approaches were proposed to alleviate this dimensionality burden, using scripts or action probability distributions, based on expert knowledge. We propose to replace expert-authored scripts with a collection of smaller parametric scripts we call heuristics and use them to pre-select actions for Monte Carlo Tree Search (MCTS). The advantages of this proposal consist of granular control of the decision space and the ability to adapt the agent's strategy in-game, all by altering the heuristics and their parameters. Experimentation results in µRTS using a proposed implementation have shown a significant performance gain over state-of-the-art agents.

IEEE CoG, 2020

The impressive performance of Monte Carlo Tree Search (MCTS) based game-playing agents in high br... more The impressive performance of Monte Carlo Tree Search (MCTS) based game-playing agents in high branching-factor domains such as Go, motivated researchers to apply and adapt MCTS to even more challenging domains. Real-time strategy (RTS) games feature a large combinatorial branching factor and a real-time aspect that pose significant challenges to a broad spectrum of AI techniques, including MCTS. Various MCTS enhancements were proposed, such as the combinatorial multi-armed bandit (CMAB) based sampling, state/action abstractions, and machine learning. In this paper, we propose to employ move pruning as a way to improve the performance of MCTS-based agents in the context of RTS games. We describe a class of possibly detrimental player-actions and propose several pruning approaches targeting it. The experimentation results in µRTS indicate that this could be a promising direction.

ICTAACS'19, 2019

Online planning is an important research area focusing on the problem of real-time decision makin... more Online planning is an important research area focusing on the problem of real-time decision making, using information extracted from the environment. The aim is to compute, at each decision point, the best decision possible that contributes to the realization of a fixed objective. Relevant application domains include robotics, control engineering and computer games. Real-time strategy (RTS) games pose considerable challenges to artificial intelligence techniques, due to their dynamic, complex and adversarial aspects, where online planning plays a prominent role. They also constitute an ideal research platform and test-bed for online planning. µRTS is an open-source AI research platform that features a minimalistic, yet complete RTS implementation, used by AI researchers for developing and testing intelligent RTS game-playing agents. The unique characteristics of µRTS helped for the emergence of interesting online adversarial planning techniques, dealing with multiple levels of abstraction. This paper presents the major µRTS online planning approaches to date, categorized by the degree of abstraction, in fully and partially observable environments.

ISPS'15, 2015

Web page classification based on topic or sentiments is a common application of web content minin... more Web page classification based on topic or sentiments is a common application of web content mining techniques. In this paper we will present a novel application based on targeting a nation. The aim is to be able to automatically distinguish websites targeting a specific nation, using both the URL and the content of a web page. In this paper we will address the issue of identifying Algerian-interest web pages using a machine learning approach.We will present the process of acquiring data for the supervised learning phase and adapting it into a usable dataset, as well as using it to construct three distinct classifiers using different parts of the data. The resulting classifiers have shown outstanding performances (up to F-score = 0.93) for such application.