Generative Models from the perspective of Continual Learning

Abstract : Which generative model is the most suitable for Continual Learning? This paper aims at evaluating and comparing generative models on disjoint sequential image generation tasks. We investigate how several models learn and forget, considering various strategies: rehearsal, regularization, generative replay and fine-tuning. We used two quantitative metrics to estimate the generation quality and memory ability. We experiment with sequential tasks on three commonly used benchmarks for Continual Learning (MNIST, Fashion MNIST). We found that among all models, the original GAN performs best and among Continual Learning strategies, gener-ative replay outperforms all other methods.
Document type :
Conference papers
Complete list of metadatas

Cited literature [14 references]  Display  Hide  Download
Contributor : Timothéee Lesort <>
Submitted on : Friday, December 21, 2018 - 11:27:29 AM
Last modification on : Wednesday, July 3, 2019 - 10:48:05 AM
Long-term archiving on : Friday, March 22, 2019 - 2:08:17 PM


Files produced by the author(s)


  • HAL Id : hal-01951954, version 1


Timothée Lesort, Hugo Caselles-Dupré, Michael Garcia-Ortiz, Jean-François Goudou, David Filliat. Generative Models from the perspective of Continual Learning. Workshop on Continual Learning, NeurIPS 2018 - Thirty-second Conference on Neural Information Processing Systems, Dec 2018, Montréal, Canada. ⟨hal-01951954⟩



Record views


Files downloads