6 citations · 18 across the 9 of their papers we have counts for
9 papers
Dynamic Sparsity: Challenging Common Sparsity Assumptions for Learning World Models in Robotic Reinforcement Learning Benchmarks
Muthukumar Pandaram, Jakob Hollenstein, David Drexel +3
The use of learned dynamics models, also known as world models, can improve the sample efficiency of reinforcement learning. Recent work suggests that the underlying causal graphs…
Unsupervised Learning of Effective Actions in Robotics
Marko Zaric, Jakob Hollenstein, Justus Piater +1
Learning actions that are relevant to decision-making and can be executed effectively is a key problem in autonomous robotics. Current state-of-the-art action representations in ro…
Colored Noise in PPO: Improved Exploration and Performance through Correlated Action Sampling
Jakob Hollenstein, Georg Martius, Justus Piater
Proximal Policy Optimization (PPO), a popular on-policy deep reinforcement learning method, employs a stochastic policy for exploration. In this paper, we propose a colored noise-b…
Scalable and Efficient Continual Learning from Demonstration via a Hypernetwork-generated Stable Dynamics Model
Sayantan Auddy, Jakob Hollenstein, Matteo Saveriano +2
Robots capable of learning from demonstration (LfD) must exhibit stability while executing learned motion skills. To be effective in the real world, they should also remember multi…
Differentiable Forward Kinematics for TensorFlow 2
Lukas Mölschl, Jakob J. Hollenstein, Justus Piater
Robotic systems are often complex and depend on the integration of a large number of software components. One important component in robotic systems provides the calculation of for…
Action Noise in Off-Policy Deep Reinforcement Learning: Impact on Exploration and Performance
Jakob Hollenstein, Sayantan Auddy, Matteo Saveriano +2
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 sc…
Continual Learning from Demonstration of Robotics Skills
Sayantan Auddy, Jakob Hollenstein, Matteo Saveriano +2
Methods for teaching motion skills to robots focus on training for a single skill at a time. Robots capable of learning from demonstration can considerably benefit from the added a…
How do Offline Measures for Exploration in Reinforcement Learning behave?
Jakob J. Hollenstein, Sayantan Auddy, Matteo Saveriano +2
Sufficient exploration is paramount for the success of a reinforcement learning agent. Yet, exploration is rarely assessed in an algorithm-independent way. We compare the behavior…
Improving the Exploration of Deep Reinforcement Learning in Continuous Domains using Planning for Policy Search
Jakob J. Hollenstein, Erwan Renaudo, Matteo Saveriano +1
Local policy search is performed by most Deep Reinforcement Learning (D-RL) methods, which increases the risk of getting trapped in a local minimum. Furthermore, the availability o…