Prof. Michael J Carbin

Professor Carbin is interested in the design of programming systems that deliver improved performance and resilience by incorporating approximate computing and self-healing. Not all parts of a program are created equal: some parts are more critical to a program's reliability and correctness than others. This observation allows us as researchers to explore aggressive program optimization and transformation methods that identify less critical regions of a program's code that can be replaced and approximated, producing a new program with better reliability or performance that still satisfies a developer's requirements.