Stochastically constrained adversarial lifelong learning

dc.contributor.authorGrant-Hagen, Mica
dc.contributor.supervisorMehta, Nishant
dc.date.accessioned2026-05-29T19:41:50Z
dc.date.available2026-05-29T19:41:50Z
dc.date.issued2026
dc.degree.departmentDepartment of Computer Science
dc.degree.levelMaster of Science MSc
dc.description.abstractLifelong learning is a setting of online learning where there is a sequence of tasks and each task is an online learning problem. The lifelong learning algorithm is split into two sections: the meta-level and the within-task level. In this thesis, we explore an online learning setting with a stochastically-constrained adversarial (SCA) assumption. The SCA assumption is that there is not a singular representation that works the best for each of the tasks in terms of minimizing expected losses, but that there is a representation that will become better than all other representations on average over all tasks after enough tasks have been complete. We show that in the single task setting, Decreasing Hedge with perturbed losses achieves best-of-both-worlds regret bounds under the SCA assumption. In the lifelong learning setting, we propose an algorithm with Decreasing Hedge in the meta-level and Squint in the within-task level. Under the SCA assumption in the meta-level and the Bernstein condition in the within-task level, this lifelong learning algorithm achieves best-of-both-worlds regret bounds.
dc.description.scholarlevelGraduate
dc.identifier.urihttps://hdl.handle.net/1828/23960
dc.languageEnglisheng
dc.language.isoen
dc.rightsAvailable to the World Wide Web
dc.subjectLifelong learning
dc.subjectOnline Learning Theory
dc.subjectMachine learning theory
dc.titleStochastically constrained adversarial lifelong learning
dc.typeThesis

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Grant-Hagen_Mica_MSc_2026.pdf
Size:
359.45 KB
Format:
Adobe Portable Document Format
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.62 KB
Format:
Item-specific license agreed upon to submission
Description: