Efficient Version Space Algorithms for Human-in-the-loop Model Development

doi:10.1145/3637443

2024-01-12

DOI：10.1145/3637443

原文求助

摘要

En 中文

When active learning (AL) is applied to help users develop a model on a large dataset through interactively presenting data instances for labeling, existing AL techniques often suffer from two main drawbacks: First, to reach high accuracy they may require the user to label hundreds of data instances, which is an onerous task for the user. Second, retrieving the next instance to label from a large dataset can be time-consuming, making it incompatible with the interactive nature of the human exploration process. To address these issues, we introduce a novel version-space-based active learner for kernel classifiers, which possesses strong theoretical guarantees on performance and efficient implementation in time and space. In addition, by leveraging additional insights obtained in the user labeling process, we can factorize the version space to perform active learning in a set of subspaces, which further reduces the user labeling effort. Evaluation results show that our algorithms significantly outperform state-of-the-art version space strategies, as well as a recent factorization-aware algorithm, for model development over large datasets.

Keyword:

Active learning

version space

kernel classifiers

hit-and-run

factorization

期刊

ACM Transactions on Knowledge Discovery from Data

IF:

4.8

论文数:

1.3K

被引数:

4.4K

机构

Ecole Polytechnique

学者数:

6.6K

论文数: 4.8K

被引数: 211

institut polytechnique de paris

学者数:

1.3W

论文数: 1.0W

被引数: 6

引用论文

Selection and Characterization of Tau Binding ᴅ-Enantiomeric Peptides with Potential for Therapy of Alzheimer Disease

PLOS ONE

IF0

2016-12-22

OAAI

Christina Dammers; Deniz Yolcu; Laura Kukuk; Dieter Willbold; Marcus Pickhardt; Eckhard Mandelkow; Anselm H. C. Horn; Heinrich Sticht; Marwa Nidal Malhis; Nadja Will; Judith Schuster; Susanne Aileen Funke

Petrography, strontium, barium and uranium concentrations, and strontium and uranium isotope ratios in speleothems as palaeoclimatic proxies: Soreq Cave, Israel

The Holocene

IF0

1999-09-01

PREAI

Avner Ayalon; Miryam Bar-Matthews; Aaron Kaufman

Morphology and position of growth cones in the developing Xenopus spinal cord

Developmental Brain Research

IF0

1982-06-01

PREAI

Ruth H. Nordlander; Marcus Singer