Variational Continual Learning

Code

• github.com/nvcuong/variational-continual-learning
  tf★ 197
• github.com/tyxe-bdl/tyxe
  pytorch★ 157
• github.com/NixGD/variational-continual-learning
  pytorch★ 0
• github.com/aml-team-2/aml-reproducability-challenge
  pytorch★ 0
• github.com/Minhchuyentoancbn/Continual-Learning
  pytorch★ 0
• github.com/pihey1995/VariationalContinualLearning
  torch★ 0
• github.com/A-Roucher/variational_continual_learning
  pytorch★ 0
• github.com/Piyush-555/VCL-in-PyTorch
  pytorch★ 0

Abstract

This paper develops variational continual learning (VCL), a simple but general framework for continual learning that fuses online variational inference (VI) and recent advances in Monte Carlo VI for neural networks. The framework can successfully train both deep discriminative models and deep generative models in complex continual learning settings where existing tasks evolve over time and entirely new tasks emerge. Experimental results show that VCL outperforms state-of-the-art continual learning methods on a variety of tasks, avoiding catastrophic forgetting in a fully automatic way.

Tasks

Reproductions