Ahmed Khlifi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Ahmed Khlifi has a total of 2 co-authors. Here are the top co-authors by number of shared publications:
- Mohamed Othmani
Computer Science Department, Faculty of Sciences, Gafsa University, Gafsa, Tunisia
1 shared publication - Monji Kherallah
University of Sfax
1 shared publication
Latest Publications
- Open Access
- 8 January 2025
- Published by MDPI AG in Preprints.org
A Novel Approach to Autonomous Driving Using DDQN-Based Deep Reinforcement Learning
Deep reinforcement learning (DRL) trains agents to make decisions by learning from rewards and penalties, using trial and error. It combines reinforcement learning with deep neural networks, enabling agents to process large datasets and learn from complex environments. DRL has achieved notable success in gaming, robotics, decision-making, etc. However, real-world applications, such as self-driving cars, face challenges due to complex state and action spaces, requiring precise control. Researchers continue to develop new algorithms to improve performance in dynamic settings. A key algorithm, Deep Q-Network (DQN), uses neural networks to approximate the Q-value function, but suffers from overestimation, leading to suboptimal outcomes. To address this, Double Deep Q-Network (DDQN) was introduced to reduce bias by separating action selection from evaluation, resulting in more stable learning. This work examines the effectiveness of DQN and DDQN in autonomous driving using the CARLA simulator, highlighting DDQN's benefits in reducing bias and enhancing policy performance.
- 8 January 2025
- Published by MDPI AG in Preprints.org
Open AccessA Novel Approach to Autonomous Driving Using DDQN-Based Deep Reinforcement Learning
Deep reinforcement learning (DRL) trains agents to make decisions by learning from rewards and penalties, using trial and error. It combines reinforcement learning with deep neural networks, enabling agents to process large datasets and learn from complex environments. DRL has achieved notable success in gaming, robotics, decision-making, etc. However, real-world applications, such as self-driving cars, face challenges due to complex state and action spaces, requiring precise control. Researchers continue to develop new algorithms to improve performance in dynamic settings. A key algorithm, Deep Q-Network (DQN), uses neural networks to approximate the Q-value function, but suffers from overestimation, leading to suboptimal outcomes. To address this, Double Deep Q-Network (DDQN) was introduced to reduce bias by separating action selection from evaluation, resulting in more stable learning. This work examines the effectiveness of DQN and DDQN in autonomous driving using the CARLA simulator, highlighting DDQN's benefits in reducing bias and enhancing policy performance.