An evolution strategy (ES) variant recently attracted significant attention due to its surprisingly good performance at optimizing neural networks in challenging deep reinforcement learning domains. It searches directly in the parameter space of neural networks by generating perturbations to the current set of parameters, checking their performance, and moving in the direction of higher reward....

The problem investigated in this paper is that of driving a car-like robot along a race track and the use of reinforcement learning to find a good control function. The reinforcement learner uses a case-based function approximator to extend the reinforcement learning paradigm to handle continuous states. The learned controller performs similar to the best control functions in both simulation an...

A neural network controller is applied to the optimal model predictive control of constrained nonlinear systems. The control law is represented by a neural network function approximator, which is trained to minimize a control-relevant cost function. The proposed procedure can be applied to construct controllers with arbitrary structures, such as optimal reduced-order controllers and decentraliz...

The concept of using ANN-like approximators for estimation of dynamic system parameters is considered. It is shown that the modular, classifier-approximator architecture offers new possibilities of realtime observation point selection for optimal ANN performance in terms of minimum parameter variance.

We investigate the approximation ability of a multilayer perceptron (MLP) network when it is extended to the complex domain. The main challenge for processing complex data with neural networks has been the lack of bounded and analytic complex nonlinear activation functions in the complex domain, as stated by Liouville's theorem. To avoid the conflict between the boundedness and the analyticity ...

In this paper, a recurrent compensatory neuro-fuzzy system (RCNFS) for identification and prediction is proposed. The compensatory-based fuzzy method uses the adaptive fuzzy operations of neuro-fuzzy systems to make fuzzy logic systems more adaptive and effective. A recurrent network is embedded in the RCNFS by adding feedback connections in the second layer, where the feedback units act as mem...

A recurrent neural network (RNN) is a universal approximator of dynamical systems, whose performance often depends on sensitive hyperparameters. Tuning them properly may be difficult and, typically, based on a trial-and-error approach. In this work, we adopt a graph-based framework to interpret and characterize internal dynamics of a class of RNNs called echo state networks (ESNs). We design pr...

How to measure the complexity of a finite set of vectors embedded in a multidimensional space? This is a non-trivial question which can be approached in many different ways. Here we suggest a set of data complexity measures using universal approximators, principal cubic complexes. Principal cubic complexes generalise the notion of principal manifolds for datasets with nontrivial topologies. The...

Multivariable nonlinear systems identification is addressed, in the following, by means of enhanced multiobjective evolutionary optimisation. The paper suggests a customised genetic programming algorithm that generates nonlinear linear in parameter models, according to a mathematical pattern that has been proven to be a universal approximator. In order to efficiently exploit the parameter wise ...