Presentation
TD-Magic: From Pictures of Timing Diagrams to Formal Specifications
DescriptionWe introduce TD-Magic, the first neuro-symbolic approach for translating a timing diagram picture (TD) to a formal specification. To overcome the lack of labelled data for training our machine-learning model, we first develop a synthetic data generator of labelled TDs. We then use object detection techniques to identify rising and failing edges, OCR to recognise the text, and image processing algorithms to capture synchronisation patterns. Finally, a semantic interpretation module analyses the extracted features and generates the associated formal specification. Our experiments over industrial TDs show high translation accuracy opening the way to more sophisticated requirements-extraction algorithms from pictures.
Event Type
Research Manuscript
TimeWednesday, July 12th11:10am - 11:25am PDT
Location3003, 3rd Floor
AI
RISC-V
AI/ML Application and Infrastructure