Alexandre M. M. Santos
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Alexandre M. M. Santos has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Artur de Oliveira da Rocha Franco
Universidade Federal do Ceará;, Brazil
1 shared publication - Creto Augusto Vidal
Federal University of Ceará, Brazil
1 shared publication - Halisson R. Rodrigues
Universidade Federal do Ceará;, Brazil
1 shared publication - Joaquim Bento Cavalcante Neto
Systems and Databases Laboratory, Federal University of Ceará
1 shared publication - Paulo Bruno S. Serafim
Gran Sasso Science Institute L'Aquila Italy
1 shared publication - Romulo F. Férrer Filho
Universidade Federal do Ceará;, Brazil
1 shared publication
Latest Publications
- Open Access
- 6 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 86-95
Unveiling the Key Features Influencing Game Agents with Different Levels of Robustness
Training agents using Deep Reinforcement Learning methods is rapidly progressing in several fields and techniques like domain randomization have been demonstrated to improve the generalization ability of these agents. However, due to the black-box nature of the models, it is not easy to understand why an action was selected from a given input. Although prior research on Explainable Artificial Intelligence presents efforts to bridge this gap, is unclear what particular input features that contribute to a model’s generalizability. This work examines the main aspects that affect the behavior of game agents with varying robustness levels. By comparing specialized and generalized agents, we investigate what are the main differences and similarities present in these models when they select an action. To achieve this goal, we trained two agents with different robustness levels and applied Explainable Artificial Intelligence methods to highlight the key features on the input screen. We employed a mixed methods analysis, which provided important quantitative results on the agents’ performance as well as qualitative insights about their behavior. We are able to show that the visualization of generalized agents tends to be more interpretable since they concentrate on the game objects, whereas specialized agents are more spread along the whole input screen. This result constitutes an important step to understanding the behavior of game agents trained using Deep Reinforcement Learning with different training procedures.
- 6 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 86-95
Open AccessUnveiling the Key Features Influencing Game Agents with Different Levels of Robustness
Training agents using Deep Reinforcement Learning methods is rapidly progressing in several fields and techniques like domain randomization have been demonstrated to improve the generalization ability of these agents. However, due to the black-box nature of the models, it is not easy to understand why an action was selected from a given input. Although prior research on Explainable Artificial Intelligence presents efforts to bridge this gap, is unclear what particular input features that contribute to a model’s generalizability. This work examines the main aspects that affect the behavior of game agents with varying robustness levels. By comparing specialized and generalized agents, we investigate what are the main differences and similarities present in these models when they select an action. To achieve this goal, we trained two agents with different robustness levels and applied Explainable Artificial Intelligence methods to highlight the key features on the input screen. We employed a mixed methods analysis, which provided important quantitative results on the agents’ performance as well as qualitative insights about their behavior. We are able to show that the visualization of generalized agents tends to be more interpretable since they concentrate on the game objects, whereas specialized agents are more spread along the whole input screen. This result constitutes an important step to understanding the behavior of game agents trained using Deep Reinforcement Learning with different training procedures.