We apply techniques from Software Engineering--including static analysis and test generation--to validate and verify properties of neural networks, such as robustness and fairness.
We help engineers build safe and reliable hardware, software, and cyber-physical systems, using a unique and promising strategy: combining inductive techniques (from the area of machine learning) with deductive techniques (from the area of mathematical logic).
Debasmita Lohar has received an award for the best talk presented at the iFM 2019 PhD Symposium.
Max Planck Researchers have authored eight papers to appear in POPL 2020, just over 10% of all papers accepted this year.
Former MPI-SWS postdoc Ori Lahav was awarded an ERC starting grant on "Verification-Aware Programming Language Concurrency Semantics".