Random path selection for incremental learning

Jathushan Rajasegaran, Munawar Hayat, Salman Khan, Fahad Shahbaz Khan, Ling Shao

Research output: A Conference proceeding or a Chapter in BookConference contributionpeer-review

53 Citations (Scopus)

Abstract

Incremental life-long learning is a main challenge towards the long-standing goal of Artificial General Intelligence. In real-life settings, learning tasks arrive in a sequence and machine learning models must continually learn to increment already acquired knowledge. Existing incremental learning approaches, fall well below the state-of-the-art cumulative models that use all training classes at once. In this paper, we propose a random path selection algorithm, called RPS-Net, that progressively chooses optimal paths for the new tasks while encouraging parameter sharing. Since the reuse of previous paths enables forward knowledge transfer, our approach requires a considerably lower computational overhead. As an added novelty, the proposed model integrates knowledge distillation and retrospection along with the path selection strategy to overcome catastrophic forgetting. In order to maintain an equilibrium between previous and newly acquired knowledge, we propose a simple controller to dynamically balance the model plasticity. Through extensive experiments, we demonstrate that the proposed method surpasses the state-of-the-art performance on incremental learning and by utilizing parallel computation this method can run in constant time with nearly the same efficiency as a conventional deep convolutional neural network.

Original languageEnglish
Title of host publicationAdvances in Neural Information Processing Systems (NeurIPS 2019)
EditorsH. Wallach, H. Larochelle, A. Beygelzimer, F. d'Alché-Buc, E. Fox, R. Garnett
Place of PublicationUnited States
PublisherAssociation for Computing Machinery (ACM)
Pages1-11
Number of pages11
Volume32
ISBN (Print)9781713807933
Publication statusPublished - 2019
Externally publishedYes
Event33rd Annual Conference on Neural Information Processing Systems, NeurIPS 2019 - Vancouver, Canada
Duration: 8 Dec 201914 Dec 2019

Publication series

NameAdvances in Neural Information Processing Systems
ISSN (Print)1049-5258

Conference

Conference33rd Annual Conference on Neural Information Processing Systems, NeurIPS 2019
Country/TerritoryCanada
CityVancouver
Period8/12/1914/12/19

Fingerprint

Dive into the research topics of 'Random path selection for incremental learning'. Together they form a unique fingerprint.

Cite this