Grammar Based Directed Testing of Machine Learning Systems
Sakshi Udeshi, Sudipta Chattopadhyay · IEEE Transactions on Software Engineering · 2019
The massive progress of machine learning has seen its application over a variety of domains in the past decade. But how do we develop a systematic, scalable and modular strategy to validate machine-learning systems? We present, to the best of our knowledge, the first approach, which provides a systematic test framework for machine-learning systems that accepts grammar-based inputs. OurOgmaapproach automatically discovers erroneous behaviours in classifiers and leverages these erroneous behaviours to improve the respective models.Ogmaleverages inherent robustness properties present in any well trained machine-learning model to direct test generation and thus, implementing a scalable test generation methodology. To evaluate ourOgmaapproach, we have tested it on three real world natural language processing (NLP) classifiers. We have found thousands of erroneous behaviours in these systems. We also compareOgmawith a random test generation approach and observe thatOgmais more effective than such random test generation by up to 489 percent.