Publications
Chosen filters:
Chosen filters:
The body as a reservoir: locomotion and sensing with linear feedback Ghent University
It is known that mass-spring nets have computational power and can be trained to reproduce oscillating patterns. In this work, we extend this idea to locomotion and sensing. We simulate systems made out of bars and springs and show that stable gaits can be maintained by these structures with only linear feedback. We then conduct a classification experiment in which the system has to distinguish terrains while maintaining an oscillatory pattern. ...
General linear thermodynamics for periodically driven systems with multiple reservoirs Hasselt University
We derive a linear thermodynamics theory for general Markov dynamics with both steady-state and time-periodic drivings. Expressions for thermodynamic quantities, such as chemical work, heat, and entropy production are obtained in terms of equilibrium probability distribution and the drivings. The entropy production is derived as a bilinear function of thermodynamic forces and the associated fluxes. We derive explicit formulae for the Onsager ...
Escherichia coli sequence type 457 is an emerging extended-spectrum-β-lactam-resistant lineage with reservoirs in wildlife and food-producing animals Vrije Universiteit Brussel
Silver gulls carry phylogenetically diverse Escherichia coli, including globally dominant extraintestinal pathogenic E. coli (ExPEC) sequence types and pandemic ExPEC-ST131 clades; however, our large-scale study (504 samples) on silver gulls nesting off the coast of New South Wales identified E. coli ST457 as the most prevalent. A phylogenetic analysis of whole-genome sequences (WGS) of 138 ST457 samples comprising 42 from gulls, 2 from ...
Supervised learning of internal models for autonomous goal-oriented robot navigation using Reservoir Computing Ghent University
In this work we propose a hierarchical architecture which constructs internal models of a robot environment for goal-oriented navigation by an imitation learning process. The proposed architecture is based on the Reservoir Computing paradigm for training Recurrent Neural Networks (RNN). It is composed of two randomly generated RNNs (called reservoirs), one for modeling the localization capability and one for learning the navigation skill. The ...
On the quantification of dynamics in Reservoir Computing Ghent University
Reservoir Computing (RC) offers a computationally efficient and well performing technique for using the temporal processing power of Recurrent Neural Networks (RNNs), while avoiding the traditional long training times and stability problems. The method is both simple and elegant: a random RNN (called the reservoir) is constructed using only a few global parameters to tune the dynamics into a desirable regime, and the dynamic response of the ...
Imitation learning of an intelligent navigation system for mobile robots using reservoir computing Ghent University
The design of an autonomous navigation system for mobile robots can be a tough task. Noisy sensors, unstructured environments and unpredictability are among the problems which must be overcome. Reservoir Computing (RC) uses a randomly created recurrent neural network (the reservoir) which functions as a temporal kernel of rich dynamics that projects the input to a high dimensional space. This projection is mapped into the desired output (only ...
An uncued brain-computer interface using reservoir computing Ghent University
Brain-Computer Interfaces are an important and promising avenue for possible next-generation assistive devices. In this article, we show how Reservoir Comput- ing – a computationally efficient way of training recurrent neural networks – com- bined with a novel feature selection algorithm based on Common Spatial Patterns can be used to drastically improve performance in an uncued motor imagery based Brain-Computer Interface (BCI). The objective ...
PyRCN : a toolbox for exploration and application of Reservoir Computing Networks Ghent University
Reservoir Computing Networks (RCNs) belong to a group of machine learning techniques that project the input space non-linearly into a high-dimensional feature space, where the underlying task can be solved linearly. Popular variants of RCNs are capable of solving complex tasks equivalently to widely used deep neural networks, but with a substantially simpler training paradigm based on linear regression. In this paper, we show how to uniformly ...
Modeling multiple autonomous robot behaviors and behavior switching with a single reservoir computing network Ghent University
Reservoir Computing (RC) uses a randomly created Recurrent Neural Network as a reservoir of rich dynamics which projects the input to a high dimensional space. These projections are mapped to the desired output using a linear output layer, which is the only part being trained by standard linear regression. In this work, RC is used for imitation learning of multiple behaviors which are generated by different controllers using an intelligent ...