MetAL: Active Semi-Supervised Learning on Graphs via Meta Learning

1 code implementation • 22 Jul 2020 • Kaushalya Madhawa, Tsuyoshi Murata

In this paper, we propose MetAL, an AL approach that selects unlabeled instances that directly improve the future performance of a classification model.

Active Learning Bilevel Optimization +1

GraphNVP: an Invertible Flow-based Model for Generating Molecular Graphs

no code implementations • 25 Sep 2019 • Kaushalya Madhawa, katsuhiko Ishiguro, Kosuke Nakago, Motoki Abe

In contrast, our model is the first invertible model for the whole graph components: both of dequantized node attributes and adjacency tensor are converted into latent vectors through two novel invertible flows.

Graph Generation Molecular Graph Generation +1

GraphNVP: An Invertible Flow Model for Generating Molecular Graphs

3 code implementations • 28 May 2019 • Kaushalya Madhawa, Katushiko Ishiguro, Kosuke Nakago, Motoki Abe

We propose GraphNVP, the first invertible, normalizing flow-based molecular graph generation model.

Graph Generation Molecular Graph Generation +1

Exploring Partially Observed Networks with Nonparametric Bandits

1 code implementation • 19 Apr 2018 • Kaushalya Madhawa, Tsuyoshi Murata

We formulate this problem as an exploration-exploitation problem and propose a novel nonparametric multi-arm bandit (MAB) algorithm for identifying which nodes to be queried.