Write a Blog >>
ICSE 2021
Mon 17 May - Sat 5 June 2021

Machine learning (ML) cloud APIs enable developers to easily incorporate learning solutions into software systems. Unfortunately, ML APIs are challenging to use correctly and efficiently, given their unique semantics, data requirements, and accuracy-performance tradeoffs. Much prior work has studied how to develop ML APIs or ML cloud services, but not how open-source applications are using ML APIs.

In this paper, we manually studied 360 representative open-source applications that use Google or AWS cloud-based ML APIs, and found 70% of these applications contain API misuses in their latest versions that degrade functional, performance, or economical quality of the software. We have generalized 8 anti-patterns based on our manual study and developed automated checkers that identify hundreds of more applications that contain ML API misuses.

Conference Day
Tue 25 May

Displayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change

19:35 - 20:55
1.5.3. API: Usage and RefactoringTechnical Track / SEIP - Software Engineering in Practice / Journal-First Papers at Blended Sessions Room 3 +12h
Chair(s): Giuseppe ScannielloUniversity of Basilicata
19:35
20m
Paper
Automatically Identifying Parameter Constraints in Complex Web APIs: A Case Study at AdyenSEIP
SEIP - Software Engineering in Practice
Henk GrentAdyen N.V., Aleksei AkimovAdyen N.V., MaurĂ­cio AnicheDelft University of Technology
Pre-print Media Attached
19:55
20m
Paper
SOAR: A Synthesis Approach for Data Science API RefactoringArtifact ReusableTechnical TrackArtifact Available
Technical Track
Ansong NiYale University, Daniel RamosCarnegie Mellon University, Aidan Z.H. YangCarnegie Mellon University, Ines LynceINESC-ID/IST, Universidade de Lisboa, Vasco ManquinhoINESC-ID/IST, Universidade de Lisboa, Ruben MartinsCarnegie Mellon University, Claire Le GouesCarnegie Mellon University
Pre-print Media Attached
20:15
20m
Paper
Studying Ad Library Integration Strategies of Top Free-to-Download AppsJournal-First
Journal-First Papers
Md AhasanuzzamanQueen's University, Safwat HassanThompson Rivers University, Ahmed E. HassanSchool of Computing, Queen's University
Link to publication DOI Pre-print Media Attached
20:35
20m
Paper
Are Machine Learning Cloud APIs Used Correctly?Artifact ReusableTechnical Track
Technical Track
Chengcheng WanUniversity of Chicago, Shicheng LiuUniversity of Chicago, Henry HoffmannUniversity of Chicago, Michael MaireUniversity of Chicago, Shan LuUniversity of Chicago
Pre-print Media Attached

Conference Day
Wed 26 May

Displayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change

07:35 - 08:55
07:35
20m
Paper
Automatically Identifying Parameter Constraints in Complex Web APIs: A Case Study at AdyenSEIP
SEIP - Software Engineering in Practice
Henk GrentAdyen N.V., Aleksei AkimovAdyen N.V., MaurĂ­cio AnicheDelft University of Technology
Pre-print Media Attached
07:55
20m
Paper
SOAR: A Synthesis Approach for Data Science API RefactoringArtifact ReusableTechnical TrackArtifact Available
Technical Track
Ansong NiYale University, Daniel RamosCarnegie Mellon University, Aidan Z.H. YangCarnegie Mellon University, Ines LynceINESC-ID/IST, Universidade de Lisboa, Vasco ManquinhoINESC-ID/IST, Universidade de Lisboa, Ruben MartinsCarnegie Mellon University, Claire Le GouesCarnegie Mellon University
Pre-print Media Attached
08:15
20m
Paper
Studying Ad Library Integration Strategies of Top Free-to-Download AppsJournal-First
Journal-First Papers
Md AhasanuzzamanQueen's University, Safwat HassanThompson Rivers University, Ahmed E. HassanSchool of Computing, Queen's University
Link to publication DOI Pre-print Media Attached
08:35
20m
Paper
Are Machine Learning Cloud APIs Used Correctly?Artifact ReusableTechnical Track
Technical Track
Chengcheng WanUniversity of Chicago, Shicheng LiuUniversity of Chicago, Henry HoffmannUniversity of Chicago, Michael MaireUniversity of Chicago, Shan LuUniversity of Chicago
Pre-print Media Attached