Self-Paced Curriculum Learning

Authors

  • Lu Jiang Carnegie Mellon University
  • Deyu Meng Xi'an Jiaotong University
  • Qian Zhao Xi'an Jiaotong University
  • Shiguang Shan Chinese Academy of Sciences
  • Alexander Hauptmann Carnegie Mellon University

DOI:

https://doi.org/10.1609/aaai.v29i1.9608

Keywords:

Self-paced Learning, Curriculum Learning, Prior Knowledge

Abstract

Curriculum learning (CL) or self-paced learning (SPL) represents a recently proposed learning regime inspired by the learning process of humans and animals that gradually proceeds from easy to more complex samples in training. The two methods share a similar conceptual learning paradigm, but differ in specific learning schemes. In CL, the curriculum is predetermined by prior knowledge, and remain fixed thereafter. Therefore, this type of method heavily relies on the quality of prior knowledge while ignoring feedback about the learner. In SPL, the curriculum is dynamically determined to adjust to the learning pace of the leaner. However, SPL is unable to deal with prior knowledge, rendering it prone to overfitting. In this paper, we discover the missing link between CL and SPL, and propose a unified framework named self-paced curriculum leaning (SPCL). SPCL is formulated as a concise optimization problem that takes into account both prior knowledge known before training and the learning progress during training. In comparison to human education, SPCL is analogous to "instructor-student-collaborative" learning mode, as opposed to "instructor-driven" in CL or "student-driven" in SPL. Empirically, we show that the advantage of SPCL on two tasks.

Downloads

Published

2015-02-21

How to Cite

Jiang, L., Meng, D., Zhao, Q., Shan, S., & Hauptmann, A. (2015). Self-Paced Curriculum Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 29(1). https://doi.org/10.1609/aaai.v29i1.9608

Issue

Section

Main Track: Novel Machine Learning Algorithms