T-ILR: a Neurosymbolic Integration for LTLf
This addresses the underexplored challenge of handling temporal logic in neurosymbolic AI for researchers in sequence-based domains, though it appears incremental as an extension of an existing algorithm.
The authors tackled the problem of integrating temporal logic specifications into deep learning for sequence-based tasks, proposing T-ILR, which improved accuracy and computational efficiency on an image sequence classification benchmark compared to the state-of-the-art.
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.