Multi-class Temporal Logic Neural Networks

arXiv:2402.12397v1 Announce Type: cross
Abstract: Time-series data can represent the behaviors of autonomous systems, such as drones and self-driving cars. The problem of binary and multi-class classification has received a lot of attention in this field. Neural networks represent a popular approach to classifying data; However, they lack interpretability, which poses a significant challenge in extracting meaningful information from them. Signal Temporal Logic (STL) is a formalism to describe the properties of timed behaviors. We propose a method that combines …

abstract arxiv attention autonomous autonomous systems binary cars challenge class classification cs.lg data driving drones interpretability logic networks neural networks popular self-driving series stat.ml systems temporal type