Pith. sign in

REVIEW

Few-Shot Learning of Compact Models via Task-Specific Meta Distillation

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2210.09922 v1 pith:E6ZJYCCY submitted 2022-10-18 cs.LG

Few-Shot Learning of Compact Models via Task-Specific Meta Distillation

classification cs.LG
keywords modelmodelsduringfew-shotmeta-learningmeta-trainingadapteddeployment
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

We consider a new problem of few-shot learning of compact models. Meta-learning is a popular approach for few-shot learning. Previous work in meta-learning typically assumes that the model architecture during meta-training is the same as the model architecture used for final deployment. In this paper, we challenge this basic assumption. For final deployment, we often need the model to be small. But small models usually do not have enough capacity to effectively adapt to new tasks. In the mean time, we often have access to the large dataset and extensive computing power during meta-training since meta-training is typically performed on a server. In this paper, we propose task-specific meta distillation that simultaneously learns two models in meta-learning: a large teacher model and a small student model. These two models are jointly learned during meta-training. Given a new task during meta-testing, the teacher model is first adapted to this task, then the adapted teacher model is used to guide the adaptation of the student model. The adapted student model is used for final deployment. We demonstrate the effectiveness of our approach in few-shot image classification using model-agnostic meta-learning (MAML). Our proposed method outperforms other alternatives on several benchmark datasets.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.