Machine Learning and Static Analysis for Building Software Verification Portfolios

Are you interested in Machine Learning and Formal Methods?

We are looking for students to work on Verifolio, our machine learning-based portfolio software verifier.
Verifolio uses static analysis and machine learning to automatically pick a suitable verification tool for a given verification task.

Currently, Verifolio is trained and evaluated on data from the Intl. Competition on Software Verification (SV-COMP) using Support Vector Machines (SVM). It outperforms any standalone software verifier in the competition. We are looking for motivated students who will

  • update Verifolio to the latest editions of SV-COMP (SV-COMP’17 & ’18)
  • facilitate further machine learning algorithms, in particular (deep) neural networks
  • advance our understanding of the trained model (e.g., for feature selection)
  • implement new static analyses to extend our feature set
  • improve our infrastructure for running experiments

 

[Contact Florian ZulegerThomas Pani]

Latest News

Jens Pagel wins Bill McCune PhD Award

We congratulate Jens Pagel for receiving the 2021 Bill McCune PhD Award in Automated Reasoning! Jens graduated in 2020; his thesis on Decision procedures for separation logic: beyond symbolic heaps (supervised by Florian Zuleger) presents his substantial contributions to the theory of formal verification and automated reasoning, and to verifying heap-manipulating programs in particular.

Continue reading

Full news archive