Enregistré dans:
Détails bibliographiques
Auteurs principaux: Jawahar, Pratik, Pierini, Maurizio
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:https://arxiv.org/abs/2603.01768
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866908887340285952
author Jawahar, Pratik
Pierini, Maurizio
author_facet Jawahar, Pratik
Pierini, Maurizio
contents Current deep learning primitives dealing with temporal dynamics suffer from a fundamental dichotomy: they are either discrete and unstable (LSTMs) \citep{pascanu_difficulty_2013}, leading to exploding or vanishing gradients; or they are continuous and dissipative (Neural ODEs) \citep{dupont_augmented_2019}, which destroy information over time to ensure stability. We propose the \textbf{Causal Hamiltonian Learning Unit} (pronounced: \textit{clue}), a novel Physics-grounded computational learning primitive. By enforcing a Relativistic Hamiltonian structure and utilizing symplectic integration, a CHLU strictly conserves phase-space volume, as an attempt to solve the memory-stability trade-off. We show that the CHLU is designed for infinite-horizon stability, as well as controllable noise filtering. We then demonstrate a CHLU's generative ability using the MNIST dataset as a proof-of-principle.
format Preprint
id arxiv_https___arxiv_org_abs_2603_01768
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle CHLU: The Causal Hamiltonian Learning Unit as a Symplectic Primitive for Deep Learning
Jawahar, Pratik
Pierini, Maurizio
Machine Learning
Artificial Intelligence
Applied Physics
Current deep learning primitives dealing with temporal dynamics suffer from a fundamental dichotomy: they are either discrete and unstable (LSTMs) \citep{pascanu_difficulty_2013}, leading to exploding or vanishing gradients; or they are continuous and dissipative (Neural ODEs) \citep{dupont_augmented_2019}, which destroy information over time to ensure stability. We propose the \textbf{Causal Hamiltonian Learning Unit} (pronounced: \textit{clue}), a novel Physics-grounded computational learning primitive. By enforcing a Relativistic Hamiltonian structure and utilizing symplectic integration, a CHLU strictly conserves phase-space volume, as an attempt to solve the memory-stability trade-off. We show that the CHLU is designed for infinite-horizon stability, as well as controllable noise filtering. We then demonstrate a CHLU's generative ability using the MNIST dataset as a proof-of-principle.
title CHLU: The Causal Hamiltonian Learning Unit as a Symplectic Primitive for Deep Learning
topic Machine Learning
Artificial Intelligence
Applied Physics
url https://arxiv.org/abs/2603.01768