(PAC-)Learning state machines from data streams: A generic strategy and an improved heuristic (Extended version)

2026-04-02

AI Summary

This work addresses the limitation of existing automata learning methods, which typically assume full access to data upfront and thus struggle in streaming settings. The authors propose a general incremental learning framework that incorporates sketch-based merging heuristics to handle incomplete prefix trees, offering the first PAC learnability guarantee for automata learning under data streams. By integrating state merging, sketch data structures, and incremental prefix tree construction, the approach achieves both theoretical correctness and substantial efficiency gains on large-scale data. Experimental results demonstrate consistent superiority over state-of-the-art methods in terms of runtime, memory consumption, and model accuracy, with validation provided through an open-source implementation on standard benchmark datasets.

Abstract

This is an extended version of our publication Learning state machines from data streams: A generic strategy and an improved heuristic, International Conference on Grammatical Inference (ICGI) 2023, Rabat, Morocco. It has been extended with a formal proof on PAC-bounds, and the discussion and analysis of a similar approach has been moved from the appendix and is now a full Section. State machines models are models that simulate the behavior of discrete event systems, capable of representing systems such as software systems, network interactions, and control systems, and have been researched extensively. The nature of most learning algorithms however is the assumption that all data be available at the beginning of the algorithm, and little research has been done in learning state machines from streaming data. In this paper, we want to close this gap further by presenting a generic method for learning state machines from data streams, as well as a merge heuristic that uses sketches to account for incomplete prefix trees. We implement our approach in an open-source state merging library and compare it with existing methods. We show the effectiveness of our approach with respect to run-time, memory consumption, and quality of results on a well known open dataset. Additionally, we provide a formal analysis of our algorithm, showing that it is capable of learning within the PAC framework, and show a theoretical improvement to increase run-time, without sacrificing correctness of the algorithm in larger sample sizes.

Problem

• state machines
• data streams
• PAC learning
• online learning
• automata learning

Innovation

• streaming data
• state machine learning
• PAC learning
• merge heuristic
• sketch-based approximation