Abstract
Deterministic Büchi automata (DBA) are useful to (probabilistic) model checking and synthesis. We survey techniques used to obtain and minimize DBAs for different classes of properties. We extend these techniques to support DBA that have generalized and transition-based acceptance (DTGBA) as they can be even smaller. Our minimization technique—a reduction to a SAT problem—synthesizes a DTGBA equivalent to the input DTGBA for any given number of states and number of acceptance sets (assuming such automaton exists). We present benchmarks using a framework that implements all these techniques.