The paper investigates how training shapes the geometry of recurrent network dynamics in Reservoir Computing (RC) networks used for temporal forecasting. It demonstrates that, in a linear continuous-time reservoir with infinite data, training data create an invariant subspace whose dimension matches the number of dominant modes. For a simplified diagonal linear reservoir, the study connects dominant eigenvalues and eigenvectors to the spectrum of a backward Dynamic Mode Decomposition matrix, providing a finite-dimensional approximation of the backward-time Koopman operator, and illustrates these phenomena through simulation while suggesting extensions to nonlinear RC.
By Noa Kaplan, Alberto Padoan, Anastasia Bizyaeva
arXiv:2607. 17858v1 Announce Type: cross Abstract: Reservoir computing exploits nonlinear dynamical systems to encode temporal inputs into high-dimensional state space representations.
By Mohab Abdalla, Damien Rontani
arXiv:2512.12767v2 Announce Type: replace-cross
Abstract: Training recurrent neuronal networks consisting of excitatory (E) and inhibitory (I) units with additive noise for working memory computation...
By Thiparat Chotibut, Oleg Evnin, Weerawit Horinouchi
arXiv:2608. 14638v1 Announce Type: new Abstract: In this paper we study autoencoders, a special class of deep neural nets (DNNs) whose performance can be characterized via their fixed points.
By Leonid Berlyand, Roman Sarapin, Yitzchak Shmalo, Victor Slavin, Sasha Sodin
arXiv:2609.13920v1 Announce Type: cross
Abstract: Persistent Gaussian perturbations have been shown to prevent asymptotic oversmoothing in recurrent Graph Neural Networks (GNNs) by ensuring a positiv...
By Mostafa Haghir Chehreghani
arXiv:2607. 17909v1 Announce Type: cross Abstract: The edge-of-chaos heuristic has long served as a guiding principle for designing reservoir computers, yet its relevance to machine performance remains elusive.
By Yao Du, Xingang Wang
arXiv:2606. 28486v1 Announce Type: cross Abstract: The emergence of low-dimensional structures in the spectra of neural network weight matrices is a common empirical feature of trained models, but the dynamical origin of this phenomenon during learning remains an open problem.
By Chanju Park, Dario Bocchi, Francesco D'Amico, Biagio Lucini, Gert Aarts
The paper investigates two strategies for incorporating heterogeneous node weights in decentralized learning: embedding the weights into local losses to use a doubly stochastic matrix, and keeping the original losses while using a λ‑induced row‑stochastic matrix. By developing a weighted Hilbert‑space framework, the authors derive tighter convergence rates and show that the row‑stochastic matrix becomes self‑adjoint, reducing penalty terms that otherwise amplify consensus error. They provide conditions under which the row‑stochastic design converges faster, even with a smaller spectral gap, and offer topology‑design guidelines based on eigenvalue comparisons.
By Bing Liu, Boao Kong, Limin Lu, Kun Yuan, Chengcheng Zhao
arXiv:2607. 14304v1 Announce Type: cross Abstract: We study sparse random geometric graphs generated by connecting pairs of high-dimensional vectors whose inner product exceeds a threshold.
By Manuel Fernandez V, Yizhe Zhu
arXiv:2607. 24420v1 Announce Type: cross Abstract: Reservoir computing has emerged as an efficient machine learning framework for predicting time series generated by dynamical systems.
By Arthur S Powanwe
arXiv:2602. 02908v2 Announce Type: replace-cross Abstract: Diffusion models trained on different, non-overlapping subsets of a dataset often produce strikingly similar outputs when given the same noise seed.
By Binxu Wang, Jacob Zavatone-Veth, Cengiz Pehlevan
arXiv:2607. 18559v1 Announce Type: cross Abstract: Gaussian graphical model selection is usually studied under independent sampling, but in many applications the data arise as a single trajectory of a dependent stochastic process.
By Vignesh Tirukkonda, Gautam Dasarathy