Luiza Lober @luizalober.bsky.social · 16/02/2026We hit the "2nd most read paper" at Journal of Physics: Complexity today! I'm quite honored that our study has attracted such interest from the scientific community, and I do hope it's useful for future studies and that it can serve as a guide for anyone interested in symbolic regression. 010
Luiza Lober @luizalober.bsky.social · 29/01/2026New open access paper just out: 𝗗𝗶𝘀𝗰𝗼𝘃𝗲𝗿𝗶𝗻𝗴 𝗲𝗾𝘂𝗮𝘁𝗶𝗼𝗻𝘀 𝗳𝗿𝗼𝗺 𝗱𝗮𝘁𝗮: 𝘀𝘆𝗺𝗯𝗼𝗹𝗶𝗰 𝗿𝗲𝗴𝗿𝗲𝘀𝘀𝗶𝗼𝗻 𝗶𝗻 𝗱𝘆𝗻𝗮𝗺𝗶𝗰𝗮𝗹 𝘀𝘆𝘀𝘁𝗲𝗺𝘀 Read the full text for free at 𝗝𝗼𝘂𝗿𝗻𝗮𝗹 𝗼𝗳 𝗣𝗵𝘆𝘀𝗶𝗰𝘀: 𝗖𝗼𝗺𝗽𝗹𝗲𝘅𝗶𝘁𝘆: iopscience.iop.org/article/10.1... Authors: @franciscorodrigues.bsky.social , Beatriz Brum and myself.iopscience.iop.orgDiscovering equations from data: symbolic regression in dynamical systemsDiscovering equations from data: symbolic regression in dynamical systems, Brum, Beatriz, Piva, Luiza Lober de Souza, Previdelli, Isolde, Rodrigues, Francisco Aparecido 031
Reposted by Luiza LoberFrancisco Rodrigues @franciscorodrigues.bsky.social · 29/08/2025Our new work on Arxiv. arxiv.org/abs/2508.20257arxiv.orgDiscovering equations from data: symbolic regression in dynamical systemsThe process of discovering equations from data lies at the heart of physics and in many other areas of research, including mathematical ecology and epidemiology. Recently, machine learning methods kno... 011
Reposted by Luiza LoberFrancisco Rodrigues @franciscorodrigues.bsky.social · 29/11/2024𝗜𝘀 𝗶𝘁 𝗽𝗼𝘀𝘀𝗶𝗯𝗹𝗲 𝘁𝗼 𝗽𝗿𝗲𝗱𝗶𝗰𝘁 𝗰𝗵𝗮𝗼𝘁𝗶𝗰 𝗱𝘆𝗻𝗮𝗺𝗶𝗰𝘀? In this new paper, we use recurrence plots to train convolutional neural networks with the task of estimating the defining control parameters of non-linear systems. arxiv.org/abs/2410.23408 with @luizalober.bsky.socialarxiv.orgPredictive Non-linear Dynamics via Neural Networks and Recurrence PlotsPredicting and characterizing diverse non-linear behaviors in dynamical systems is a complex challenge, especially due to the inherently presence of chaotic dynamics. Current forecasting methods are r... 0153
Reposted by Luiza LoberFrancisco Rodrigues @franciscorodrigues.bsky.social · 01/11/2024Our new preprint on arxiv is out. We present a methodology that leverages recurrence plots as input data for training convolutional neural networks to estimate the control parameters governing two distinct nonlinear systems: (i) the Logistic map and (ii) the Standard map. arxiv.org/abs/2410.23408arxiv.orgPredictive Non-linear Dynamics via Neural Networks and Recurrence PlotsPredicting and characterizing diverse non-linear behaviors in dynamical systems is a complex challenge, especially due to the inherently presence of chaotic dynamics. Current forecasting methods are r... 062