Residual policy learning facilitates efficient model-free autonomous racing

R Zhang, J Hou, G Chen, Z Li, J Chen… - IEEE Robotics and …, 2022‏ - ieeexplore.ieee.org
Motion planning for autonomous racing is a challenging task due to the safety requirement
while driving aggressively. Most previous solutions utilize the prior information or depend on …

Action noise in off-policy deep reinforcement learning: Impact on exploration and performance

J Hollenstein, S Auddy, M Saveriano… - arxiv preprint arxiv …, 2022‏ - arxiv.org
Many Deep Reinforcement Learning (D-RL) algorithms rely on simple forms of exploration
such as the additive action noise often used in continuous control domains. Typically, the …

Solving continuous control via q-learning

T Seyde, P Werner, W Schwarting… - arxiv preprint arxiv …, 2022‏ - arxiv.org
While there has been substantial success for solving continuous control with actor-critic
methods, simpler critic-only methods such as Q-learning find limited application in the …

Latent imagination facilitates zero-shot transfer in autonomous racing

A Brunnbauer, L Berducci… - … on Robotics and …, 2022‏ - ieeexplore.ieee.org
World models learn behaviors in a latent imagination space to enhance the sample-
efficiency of deep reinforcement learning (RL) algorithms. While learning world models for …

Continuous control with coarse-to-fine reinforcement learning

Y Seo, J Uruç, S James - arxiv preprint arxiv:2407.07787, 2024‏ - arxiv.org
Despite recent advances in improving the sample-efficiency of reinforcement learning (RL)
algorithms, designing an RL algorithm that can be practically deployed in real-world …

Continuous control with action quantization from demonstrations

R Dadashi, L Hussenot, D Vincent, S Girgin… - arxiv preprint arxiv …, 2021‏ - arxiv.org
In this paper, we propose a novel Reinforcement Learning (RL) framework for problems with
continuous action spaces: Action Quantization from Demonstrations (AQuaDem). The …

Growing Q-networks: Solving continuous control tasks with adaptive control resolution

T Seyde, P Werner, W Schwarting… - … Annual Learning for …, 2024‏ - proceedings.mlr.press
Recent reinforcement learning approaches have shown surprisingly strong capabilities of
bang-bang policies for solving continuous control benchmarks. The underlying coarse …

Reinforcement learning with simple sequence priors

T Saanum, N Éltető, P Dayan… - Advances in Neural …, 2023‏ - proceedings.neurips.cc
In reinforcement learning (RL), simplicity is typically quantified on an action-by-action basis--
but this timescale ignores temporal regularities, like repetitions, often present in sequential …

Distributional reinforcement learning-based energy arbitrage strategies in imbalance settlement mechanism

SSK Madahi, B Claessens, C Develder - Journal of Energy Storage, 2024‏ - Elsevier
Growth in the penetration of renewable energy sources makes supply more uncertain and
leads to an increase in the system imbalance. This trend, together with the single imbalance …

Geometric fabrics: a safe guiding medium for policy learning

K Van Wyk, A Handa, V Makoviychuk… - … on Robotics and …, 2024‏ - ieeexplore.ieee.org
Robotics policies are always subjected to complex, second order dynamics that entangle
their actions with resulting states. In reinforcement learning (RL) contexts, policies have the …