[1905.11063] Dataset2Vec: Learning Dataset Meta-Features

Meta-learning, or learning to learn, is a machine learning approach that utilizes prior learning experiences to expedite the learning process on unseen tasks. As a data-driven approach, meta-learning requires meta-features that represent the primary learning tasks or datasets, and are estimated traditonally as engineered dataset statistics that require expert domain knowledge tailored for every meta-task. In this paper, first, we propose a meta-feature extractor called Dataset2Vec that combines the versatility of engineered dataset meta-features with the expressivity of meta-features learned by deep neural networks. Primary learning tasks or datasets are represented as hierarchical sets, i.e., as a set of sets, esp. as a set of predictor/target pairs, and then a DeepSet architecture is employed to regress meta-features on them. Second, we propose a novel auxiliary meta-learning task with abundant data called dataset similarity learning that aims to predict if two batches stem from the

Keywords: dataset
Date: 2021/01/12 23:21

Related Entries

Read more Research projects using Chainer
0 users, 0 mentions 2018/04/22 03:40
Read more 勾配法(Gradient Descent)における学習率(learning rate)の設定と値の収束(convergence)について - YouTube
0 users, 1 mentions 2020/12/04 11:22
Read more 効果検証入門から見直す 「データサイエンス」 - Speaker Deck
0 users, 0 mentions 2020/12/06 08:21
Read more Japan.R 2020 - YouTube
0 users, 0 mentions 2020/12/19 11:21
Read more Seminar Report: The 1st STAIR Lab ST Seminar: Dr. Hiroshi Unno, “Program Verification and Synthesis ...
0 users, 1 mentions 2020/12/21 23:22