Researchers have developed SYMAR, a new method for verifying probabilistic timed automata (PTAs) that significantly outperforms existing techniques. SYMAR makes the value function symbolic, representing probability bounds as functions of sub-regions within clock regions, rather than splitting regions until uniform. This approach, guided by heuristic forward search on an And-Or graph, confines computational effort to the property-relevant fragment of the system. On benchmarks for autonomous systems, SYMAR achieved up to 108x speedup and solved problems that previously exhausted computational limits. AI
IMPACT SYMAR's efficiency in verifying complex systems could accelerate the development and deployment of reliable autonomous agents.
RANK_REASON Academic paper detailing a new method for verifying probabilistic timed automata. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →