Výsledky vyhledávání - "Wachel, Paweł"

Report

Kernel-based learning with guarantees for multi-agent applications

Autor: Kowalczyk, Krzysztof, Wachel, Paweł, Rojas, Cristian R.

This paper addresses a kernel-based learning problem for a network of agents locally observing a latent multidimensional, nonlinear phenomenon in a noisy environment. We propose a learning algorithm that requires only mild a priori knowledge about th

Externí odkaz: http://arxiv.org/abs/2404.09708

Zobrazit plný text záznamu

Report

A computationally lightweight safe learning algorithm

Autor: Baumann, Dominik, Kowalczyk, Krzysztof, Tiels, Koen, Wachel, Paweł

Safety is an essential asset when learning control policies for physical systems, as violating safety constraints during training can lead to expensive hardware damage. In response to this need, the field of safe learning has emerged with algorithms

Externí odkaz: http://arxiv.org/abs/2309.03672

Zobrazit plný text záznamu

Report

Orthogonal Transforms in Neural Networks Amount to Effective Regularization

Autor: Zając, Krzysztof, Sopot, Wojciech, Wachel, Paweł

We consider applications of neural networks in nonlinear system identification and formulate a hypothesis that adjusting general network structure by incorporating frequency information or other known orthogonal transform, should result in an efficie

Externí odkaz: http://arxiv.org/abs/2305.06344

Zobrazit plný text záznamu

Report

Decentralized diffusion-based learning under non-parametric limited prior knowledge

Autor: Wachel, Paweł, Kowalczyk, Krzysztof, Rojas, Cristian R.

We study the problem of diffusion-based network learning of a nonlinear phenomenon, $m$, from local agents' measurements collected in a noisy environment. For a decentralized network and information spreading merely between directly neighboring nodes

Externí odkaz: http://arxiv.org/abs/2305.03295

Zobrazit plný text záznamu

Report

On state-space representations of general discrete-time dynamical systems

Autor: Rojas, Cristian R., Wachel, Pawel

In this paper we establish that every (deterministic) non-autonomous, discrete-time, causal, time invariant system has a state-space representation, and discuss its minimality.
Comment: Accepted to the IEEE Transactions on Automatic Control

Externí odkaz: http://arxiv.org/abs/2205.03366

Zobrazit plný text záznamu