Transfer '13: Transfer Learning in Software Engineering Workshop (Palo Alto)

https://promisedata.or... (https://promisedata.org/transfer/13/)

The International Workshop on Transfer Learning in Software Engineering
Silicon Valley, California November 12th, 2013

In association with the 28th International Conference on Automated Software Engineering

About the Workshop

Why is generality so hard to find in software engineering? Do such general principles exist (and we just have not found them yet)? Or are we doomed to a perpetual revision of all our coding and management practices for each new project? On this issue, we can identify two feuding schools of thought. Globalists and localists use different strategies for learning best practices. Globalists learn lessons once from all data then reuse those lessons at multiple sites, whereas localists learn best practices many times from local data and use them only at a single site (which implies that project managers must devote their scant resources to a "local lessons team" that pursues best local practices).

This workshop will focus on the following issue: Whose strategy is best for finding project-specific best practices: the globalists or the localists? Or perhaps it is best to combine both approaches. It is certainly important to learn best practices from past projects. However, it is just as important to know how to transfer and adapt that experience to current projects. No project is exactly like previous projects -- hence, the trick is to find which parts of the past are most relevant and can be transferred into the current project.

Topics

Topics of interest are theoretical foundations and practical approaches related, but not limited, to the integration of the transfer learning and software engineering. While our focus in on automatic methods, we are also interested in hybrid human/computer methods.

Paper Submission

We invite submissions in either of the following forms:
-- Research papers presenting novel approaches with new research results (maximum 8 pages including references).
-- Position papers describing new ideas not yet fully developed or preliminary tool support (maximum 4 pages including references).

Submitted papers must be written in English, should not have been submitted for review or published elsewhere, and must conform to the ACM SIG proceedings templates. Submissions are accepted via Easy Chair. Submit your paper here.

Proceedings & Special Issue

Accepted papers will be published in the workshop proceedings and available electronically via ACM Digital Library within its International Conference Proceedings Series. Authors of best papers will be invited to submit a revised and extended version of their papers to a journal special issue, TBD.