analiza aiqbench
This paper proposes a classification method in which a Neural ODE is trained so that each class corresponds to a planted fixed point, and inputs are classified by the basin of attraction they flow into. It is a theoretical and methodological contribution to continuous-depth models, and its practical value depends on whether the approach matches standard classifiers on established benchmarks at comparable cost. Readers interested in dynamical-systems views of neural networks should check the reported experiments before drawing conclusions.
Approximating velocity fields with planted point attractors via Neural-ODE for classification purposes
arXiv:2606.23550v3 Announce Type: replace-cross Abstract: In this work, Neural ODEs equipped with a curated collection of equilibrium points have been successfully employed for classification tasks. The planted fixed points serve as indicators for the target classes, while the velocity field, leveraging the universal approximation capabilities of the architecture, shapes the dynamical landscape. This process defines the basins of attraction of the trained model, effectively directing each input (provided as an initial condition) toward its corresponding destination target.
Abstract:In this work, Neural ODEs equipped with a curated collection of equilibrium points have been successfully employed for classification tasks. The planted fixed points serve as indicators for the target classes, while the velocity field, leveraging the universal approximation capabilities of the architecture, shapes the dynamical landscape. This process defines the basins of attraction of the trained model, effectively directing each input (provided as an initial condition) toward its corresponding destination target.
Disordered Systems and Neural Networks (cond-mat.dis-nn); Machine Learning (cs.LG)
arXiv:2606.23550 [cond-mat.dis-nn]
(or arXiv:2606.23550v3 [cond-mat.dis-nn] for this version)
https://doi.org/10.48550/arXiv.2606.23550 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 170 cuvinte
Subiecte conexe
SW-KAN: Kolmogorov-Arnold Networks with Stieltjes-Wigert q-Orthogonal Polynomials
arXiv · 2 oct. 2026
Supervised Learning with Quantum Hamiltonian Embeddings
arXiv · 21 oct. 2026
Scaling Legal AI: Benchmarking Mamba and Transformers for Statutory Classification and Case Law Retrieval
arXiv · 8 oct. 2026
Gaussian Approximation of Besov Functions for Logistic Classification on Manifolds
arXiv · 21 oct. 2026