Â鶹ÊÓƵ

Elizabeth Dinella’s research interests are in Software Engineering and Machine Learning, focusing on the integration of symbolic program analysis and neural techniques to improve software correctness. Methods to productively write secure and correct programs are imperative in our society, which is underpinned by software systems. Software bugs and vulnerabilities are commonplace and can have devastating impacts. Currently, existing techniques to analyze programs for faults have fundamental limitations, preventing widespread deployment. Elizabeth seeks to address their shortcomings with the strengths of neural models. Ultimately, her research objective is to create program analysis techniques that are effective and reliable through Cooperative Program Reasoning and Neural Modeling. Elizabeth's foundational work on AI for software engineering has been highly cited, patented, and deployed in industrial systems.

Elizabeth Dinella’s teaching approach emphasizes bridging theory and practice and breaking down complex concepts in an accessible way. In many computer science courses, understanding how abstract, mathematical concepts apply to real-world scenarios can be challenging. Elizabeth strives to connect high-level theory with hands-on experiences.