paper-with-me

홈 › Papers

Do Data-based Curricula Work?

2022-05-01 · insights (ACL) 2022 5 · Maxim Surkov, Vladislav Mosin, Ivan Yamshchikov

Current state-of-the-art NLP systems use large neural networks that require extensive computational resources for training. Inspired by human knowledge acquisition, researchers have proposed curriculum learning - sequencing tasks (task-based curricula) or ordering and sampling the datasets (data-based curricula) that facilitate training. This work investigates the benefits of data-based curriculum learning for large language models such as BERT and T5. We experiment with various curricula based on complexity measures and different sampling strategies. Extensive experiments on several NLP tasks show that curricula based on various complexity measures rarely have any benefits, while random sampling performs either as well or better than curricula.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Do Data-based Curricula Work?

2021-12-13 · Maxim K. Surkov, Vladislav D. Mosin, Ivan P. Yamshchikov

Current state-of-the-art NLP systems use large neural networks that require lots of computational resources for training. Inspired by human knowledge acquisition, researchers have proposed curriculum learning, - sequenci…

HuCurl: Human-induced Curriculum Discovery

2023-07-14 · Mohamed Elgaar, Hadi Amiri

We introduce the problem of curriculum discovery and describe a curriculum learning framework capable of discovering effective curricula in a curriculum space based on prior knowledge about sample difficulty. Using annot…

Integrating Curricula with Replays: Its Effects on Continual Learning

2023-07-08 · Ren Jie Tee, Mengmi Zhang

Humans engage in learning and reviewing processes with curricula when acquiring new skills or knowledge. This human learning behavior has inspired the integration of curricula with replay methods in continual learning ag…

Continual LearningTransfer Learning

Pretraining Curricula Enable Selective Fine-tuning

2026-07-06 · Sebastian A. Bruijns, Jirko Rubruck, Mia H. Whitefield, Kai J. Sandbrink 외 arxiv

Transformers follow implicit curricula whereby some tasks are learned before others. However, how explicit pretraining curricula influence learning, generalization, and the selectivity of fine-tuning is unclear. This is …

Curriculum learning as a tool to uncover learning principles in the brain

2021-09-29 · ICLR 2022 4 · Daniel R. Kepple, Rainer Engelken, Kanaka Rajan

We present a novel approach to use curricula to identify principles by which a system learns. Previous work in curriculum learning has focused on how curricula can be designed to improve learning of a model on particular…