Saved in:
Bibliographic Details
Main Authors: Umili, Elena, Argenziano, Francesco, Capobianco, Roberto
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.03486
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914304129761280
author Umili, Elena
Argenziano, Francesco
Capobianco, Roberto
author_facet Umili, Elena
Argenziano, Francesco
Capobianco, Roberto
contents Integrating logical knowledge into deep neural network training is still a hard challenge, especially for sequential or temporally extended domains involving subsymbolic observations. To address this problem, we propose DeepDFA, a neurosymbolic framework that integrates high-level temporal logic - expressed as Deterministic Finite Automata (DFA) or Moore Machines - into neural architectures. DeepDFA models temporal rules as continuous, differentiable layers, enabling symbolic knowledge injection into subsymbolic domains. We demonstrate how DeepDFA can be used in two key settings: (i) static image sequence classification, and (ii) policy learning in interactive non-Markovian environments. Across extensive experiments, DeepDFA outperforms traditional deep learning models (e.g., LSTMs, GRUs, Transformers) and novel neuro-symbolic systems, achieving state-of-the-art results in temporal knowledge integration. These results highlight the potential of DeepDFA to bridge subsymbolic learning and symbolic reasoning in sequential tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2602_03486
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle DeepDFA: Injecting Temporal Logic in Deep Learning for Sequential Subsymbolic Applications
Umili, Elena
Argenziano, Francesco
Capobianco, Roberto
Machine Learning
Artificial Intelligence
Integrating logical knowledge into deep neural network training is still a hard challenge, especially for sequential or temporally extended domains involving subsymbolic observations. To address this problem, we propose DeepDFA, a neurosymbolic framework that integrates high-level temporal logic - expressed as Deterministic Finite Automata (DFA) or Moore Machines - into neural architectures. DeepDFA models temporal rules as continuous, differentiable layers, enabling symbolic knowledge injection into subsymbolic domains. We demonstrate how DeepDFA can be used in two key settings: (i) static image sequence classification, and (ii) policy learning in interactive non-Markovian environments. Across extensive experiments, DeepDFA outperforms traditional deep learning models (e.g., LSTMs, GRUs, Transformers) and novel neuro-symbolic systems, achieving state-of-the-art results in temporal knowledge integration. These results highlight the potential of DeepDFA to bridge subsymbolic learning and symbolic reasoning in sequential tasks.
title DeepDFA: Injecting Temporal Logic in Deep Learning for Sequential Subsymbolic Applications
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2602.03486