Ranking Neural Checkpoints

Li, Yandong; Jia, Xuhui; Sang, Ruoxin; Zhu, Yukun; Green, Bradley; Wang, Liqiang; Gong, Boqing

doi:10.48550/arxiv.2011.11200

Search citation statements

Order By: Relevance

Paper Sections

Select...

Introduction1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2021

Publication Types

Select...

Other1

Relationship

Self Cite0

Independent1

Authors

Journals

Cited by 1 publication

(1 citation statement)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Each of these methods has its own pros and cons. However, these studies mainly aim at interpreting the performance of transfer learning [9,10]. They mainly attack the question, "For a target task, which source task is the most relevant?".…”

Section: Introductionmentioning

confidence: 99%

Exploiting a Zoo of Checkpoints for Unseen Tasks

Huang¹,

Qiu²,

Church³

2021

Preprint

View full text Add to dashboard Cite

There are so many models in the literature that it is difficult for practitioners to decide which combinations are likely to be effective for a new task. This paper attempts to address this question by capturing relationships among checkpoints published on the web. We model the space of tasks as a Gaussian process. The covariance can be estimated from checkpoints and unlabeled probing data. With the Gaussian process, we can identify representative checkpoints by a maximum mutual information criterion. This objective is submodular. A greedy method identifies representatives that are likely to "cover" the task space. These representatives generalize to new tasks with superior performance. Empirical evidence is provided for applications from both computational linguistics as well as computer vision 1 .

show abstract

Section: Introductionmentioning

confidence: 99%