5 papers
Higher Resolution, Better Generalization: Unlocking Visual Scaling in Deep Reinforcement Learning
Raphael Trumpp, Ãmer Veysel ÃaÄatan, Ömer Veysel Çağatan +3
Pixel-based deep reinforcement learning agents are typically trained on heavily downsampled visual observations, a convention inherited from early benchmarks rather than grounded i…
Efficient Real-World Autonomous Racing via Attenuated Residual Policy Optimization
Raphael Trumpp, Denis Hoornaert, Mirco Theile +1
Residual policy learning (RPL), in which a learned policy refines a static base policy using deep reinforcement learning (DRL), has shown strong performance across various robotic…
Impoola: The Power of Average Pooling for Image-Based Deep Reinforcement Learning
Raphael Trumpp, Ansgar Schäfftlein, Mirco Theile +1
As image-based deep reinforcement learning tackles more challenging tasks, increasing model size has become an important factor in improving performance. Recent studies achieved th…
From Marginal to Joint Predictions: Evaluating Scene-Consistent Trajectory Prediction Approaches for Automated Driving
Fabian Konstantinidis, Ariel Dallari Guerreiro, Raphael Trumpp +4
Accurate motion prediction of surrounding traffic participants is crucial for the safe and efficient operation of automated vehicles in dynamic environments. Marginal prediction mo…
Action Mapping for Reinforcement Learning in Continuous Environments with Constraints
Mirco Theile, Lukas Dirnberger, Raphael Trumpp +2
Deep reinforcement learning (DRL) has had success across various domains, but applying it to environments with constraints remains challenging due to poor sample efficiency and slo…