T-ILR: a Neurosymbolic Integration for LTLf

TL;DR

T-ILR introduces a neurosymbolic framework that directly incorporates Linear Temporal Logic over finite traces into deep learning models, enhancing sequence classification accuracy and efficiency.

Contribution

It extends the ILR algorithm to handle LTLf specifications directly, bypassing the need for explicit automaton representations in temporal neurosymbolic integration.

Findings

Improved accuracy on temporal sequence classification benchmark

Enhanced computational efficiency over existing methods

Effective integration of LTLf into deep learning architectures

Abstract

State-of-the-art approaches for integrating symbolic knowledge with deep learning architectures have demonstrated promising results in static domains. However, methods to handle temporal logic specifications remain underexplored. The only existing approach relies on an explicit representation of a finite-state automaton corresponding to the temporal specification. Instead, we aim at proposing a neurosymbolic framework designed to incorporate temporal logic specifications, expressed in Linear Temporal Logic over finite traces (LTLf), directly into deep learning architectures for sequence-based tasks. We extend the Iterative Local Refinement (ILR) neurosymbolic algorithm, leveraging the recent introduction of fuzzy LTLf interpretations. We name this proposed method Temporal Iterative Local Refinement (T-ILR). We assess T-ILR on an existing benchmark for temporal neurosymbolic architectures, consisting of the classification of image sequences in the presence of temporal knowledge. The results demonstrate improved accuracy and computational efficiency compared to the state-of-the-art method.