Deep Learning for Code Generation: Status quo and challenges

Speaker: Professor Zhi Jin
         Peking University

Title:  "Deep Learning for Code Generation: Status quo and challenges"

Date:   Thursday, 13 July 2023

Time:   2:00pm - 3:00pm

Venue:  Room 5506 (via lift 25/26), HKUST


Abstract:

Over the past decade, we have witnessed a whole new era of automated code
generation due to the power of deep learning technology. This talk will
sort out the development of deep code generation by formalizing the
pipeline and code generation process in general. It also categorizes
existing solutions based on taxonomy from the perspective of architecture,
model-agnostic enhancement strategies, metrics, and tasks respectively.
Finally, this talk will also discuss the challenges posed by the currently
dominant large models and presents some suggestions about the plausible
directions for future research.


***********************
Biography:

Zhi Jin is professor of computer science at Peking University. She is
currently the deputy director of Key Lab of High Confidence Software
Technologies (PKU), Ministry of Education. Her main research interest is
AI for SE, with a long-term focus on deep learning based code
comprehension and domain knowledge-led requirements engineering. She has
published over 200 scientific articles in refereed international journals,
such as IEEE TKDE, TSE, ACM TOSEM, and TCPS, and high rank
conferences, such as ICSE, FSE, ASE and RE. She has co-authored five books
and has held more than 30 approved invention patents. She is five times
recipient of ACM SIGSOFT Distinguished Paper Awards. Prof. Jin serves as
Executive Editor-in-Chief of CJoS, an Associate Editor of IEEE TSE, IEEE
TR, and ACM TAAS, and serves on the Editorial Board of REJ and ESE. She
was the director of CCF TCSE (2016-2019) and currently the director of CCF
TCSS (2020-2023). Prof. Jin is a Fellow of CCF (China Computer
Federation), and a Fellow of IEEE (Institute of Electrical and Electronics
Engineers).