16 citations · 62 across the 13 of their papers we have counts for
16 papers
In CEM, a World Model Is Also a Proposal Mechanism
Oliver Obst, Frieder Stolzenburg
The cross-entropy method (CEM) uses world-model scores to select action sequences and fit the distribution sampled in its next iteration. A scoring error can therefore change both…
Linear Recurrent Memory Suffices to Distil a World-Model Policy for Robot Air Hockey
F. Olivia Fan, Oliver Obst
Does memory-dependent control need nonlinear recurrent dynamics? We study simulated air-hockey defence under temporary loss of puck tracking. A DreamerV3 teacher outperforms a memo…
Analyzing Echo-state Networks Using Fractal Dimension
Norbert Michael Mayer, Oliver Obst
This work joins aspects of reservoir optimization, information-theoretic optimal encoding, and at its center fractal analysis. We build on the observation that, due to the recursiv…
A hydraulic model outperforms work-balance models for predicting recovery kinetics from intermittent exercise
Fabian C. Weigend, David C. Clarke, Oliver Obst +1
Data Science advances in sports commonly involve "big data", i.e., large sport-related data sets. However, such big data sets are not always available, necessitating specialized mo…
Information Maximizing Exploration with a Latent Dynamics Model
Trevor Barron, Oliver Obst, Heni Ben Amor
All reinforcement learning algorithms must handle the trade-off between exploration and exploitation. Many state-of-the-art deep reinforcement learning methods use noise in the act…
Efficient Time-Series Approximation with Linear Recurrent Neural Networks: Architecture Learning and Predictive Power
Frieder Stolzenburg, Sandra Litz, Olivia Michael +1
Recurrent neural networks are a powerful means to cope with time series. We show how autoregressive linear, i.e., linearly activated recurrent neural networks (LRNNs) can approxima…