Pavel Smrz

Pavel Smrz is identified in the provided evidence as a co-author of the arXiv paper “Automation of Processor Verification Using Recurrent Neural Networks,” submitted on 2018-03-06 with Martin Fajcik and Marcela Zachariasova.

Technical work

The cited paper addresses simulation-based processor verification, where stimuli are generated using pseudorandom generators, applied to processor inputs, and evaluated through functional coverage. The paper proposes dynamically altering pseudorandom-generator constraints using a recurrent neural network that receives coverage feedback from simulation of the design under verification.

For demonstration, the work used processors provided by Codasip, noting that their coverage state spaces were large enough and varied across processor kinds. The abstract reports that the approach achieved coverage closure sooner and isolated a small high-coverage stimulus set suitable for regression tests.

Verification status

Only one sourced technical publication is available in the provided evidence. No sourced biographical details, institutional affiliation, education history, or broader publication record are available here.