Prem Devanbu holds a B.Tech from IIT Madras, and a Ph.D from Rutgers University. After circa 20 years at Bell Labs, he joined UC Davis, where he is now a Distinguished Research Professor of Computer Science. He works in Empirical Software Engineering, and AI for SE, specifically exploiting "big data" available in software repositories to support software development. He is a winner of  the ACM SIGSOFT Outstanding Research Award <https://www.sigsoft.org/awards/outstandingResearchAward.html> (2021), and the Alexander von Humboldt Research Award <https://www.humboldt-foundation.de/en/apply/sponsorship-programmes/humboldt-research-award> (2022), and several best-paper,  most-influential paper, and test-of-time awards. He is a Fellow of the ACM.

AG 1, AG 2, AG 3, INET, AG 4, AG 5, D6, SWS, RG1, MMCI   Info

Public Audience

English

60 Minutes

Saarbrücken

After discovering, back in 2011, that Language Models are useful for modeling repetitive patterns in source code (c.f. The "Naturalness" of software <https://dl.acm.org/doi/10.5555/2337223.2337322>), and exploring some applications thereof, more recently (since about 2019) our group at UC Davis has focused on the observation that Software, as usually written, is bimodal, admitting both the well-known formal, deterministic semantics (mostly for machines) and probabilistic, noisy semantics (for humans). This bimodality property affords both new approaches to software tool construction (using machine-learning) and new ways of studying human code reading. In this talk, I'll give an overview of the Naturalness/Bimodality program, and some recent work we have done on calibrating the quality of code produced by large language models, and also on "bimodal prompting".

--

Please contact the office team for link information

Claudia Richter

+49 681 9303 9103

--email hidden

Yes

Kaiserslautern

G26

111

Zoom

passcode not visible

logged in users only