Abstract
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 same dataset or different ones. In an experiment on a large-scale hyperparameter optimization task for 120 UCI datasets with varying schemas as a meta-learning task, we show that the meta-features of Dataset2Vec outperform the expert engineered meta-features and thus demonstrate the usefulness of learned meta-features for datasets with varying schemas for the first time.
Highlights
Meta-learning, or learning to learn, refers to any learning approach that systematically makes use of prior learning experiences to accelerate training on unseen tasks or datasets (Vanschoren 2018)
A way more simple, unsupervised plausibility argument for the usefulness of the extracted meta-features is depicted in Fig. 1 showing a 2D embedding of the meta-features of 2000 synthetic classification toy datasets of three different types computed by (a) two sets of engineered dataset meta-features: MF1 (Wistuba et al 2016) and MF2 (Feurer et al 2015); (b) a stateof-the-art model based on variational autoencoders, the Neural Statistician (Edwards and Storkey 2017b), and (c) the proposed meta-feature extractor Dataset2Vec
We show experimentally that using the meta-features extracted through Dataset2Vec for the hyperparameter optimization meta-task outperforms the use of engineered meta-features designed for this meta-task
Summary
Meta-learning, or learning to learn, refers to any learning approach that systematically makes use of prior learning experiences to accelerate training on unseen tasks or datasets (Vanschoren 2018). We design a novel meta-feature extractor called Dataset2Vec, that learns metafeatures from (tabular) datasets of a varying number of instances, predictors, or targets. A way more simple, unsupervised plausibility argument for the usefulness of the extracted meta-features is depicted in Fig. 1 showing a 2D embedding of the meta-features of 2000 synthetic classification toy datasets of three different types (circles/moon/blobs) computed by (a) two sets of engineered dataset meta-features: MF1 (Wistuba et al 2016) and MF2 (Feurer et al 2015) (see Table 3); (b) a stateof-the-art model based on variational autoencoders, the Neural Statistician (Edwards and Storkey 2017b), and (c) the proposed meta-feature extractor Dataset2Vec. For the 2D embedding, multi-dimensional scaling has been applied (Borg and Groenen 2003) on these meta-features. We show experimentally that using the meta-features extracted through Dataset2Vec for the hyperparameter optimization meta-task outperforms the use of engineered meta-features designed for this meta-task
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have