Learning Universally Quantified Invariants of Linear Data Structures

Provided by: University of Illinois at Urbana Champaign
Topic: Big Data
Format: PDF
The authors propose a new automaton model, called quantified data automata over words that can model quantified invariants over linear data structures, and build poly-time active learning algorithms for them, where the learner is allowed to query the teacher with membership and equivalence queries. In order to express invariants in decidable logics, they invent a decidable subclass of QDAs, called elastic QDAs, and prove that every QDA has a unique minimally-over-approximating elastic QDA. They then give an application of these theoretically sound and efficient active learning algorithms in a passive learning framework and show that they can efficiently learn quantified linear data structure invariants from samples obtained from dynamic runs for a large class of programs.

Find By Topic