Martin Fajcik

Martin Fajcik is documented in the available evidence as an author of the computer-science paper “Automation of Processor Verification Using Recurrent Neural Networks.” arXiv metadata lists the paper’s authors as Martin Fajcik, Marcela Zachariasova, and Pavel Smrz, with an online date of 2018-03-06 and arXiv identifier 1803.09810.[1]

Documented work

“Automation of Processor Verification Using Recurrent Neural Networks”

The paper addresses simulation-based processor verification. Its abstract describes a workflow in which pseudorandom generators produce stimuli, the stimuli are applied to processor inputs, and achieved functional coverage is monitored to determine verification completeness.[2]

The proposed technique dynamically alters constraints for a pseudorandom generator using a recurrent neural network that receives coverage feedback from simulation of the design under verification.[3] For demonstration, the authors used processors provided by Codasip, noting that their coverage state spaces were reasonably large and varied across processor types.[4]

According to the paper abstract, the experimental results showed faster coverage closure and the ability to isolate a small set of high-coverage stimuli suitable for regression tests.[5]

Scope of available evidence

The provided evidence supports Martin Fajcik’s authorship of the arXiv paper above and its technical subject matter. It does not provide verified biographical details such as institutional affiliation, education, nationality, or career history.