Learning Dynamic Latent Spaces for Lifelong Generative Modelling

TY - GEN

T1 - Learning Dynamic Latent Spaces for Lifelong Generative Modelling

AU - Ye, Fei

AU - Bors, Adrian Gheorghe

N1 - This is an author-produced version of the published paper. Uploaded in accordance with the publisher’s self-archiving policy. Further copying may not be permitted; contact the publisher for details

PY - 2023

Y1 - 2023

N2 - Task Free Continual Learning (TFCL) aims to capture novel concepts from non-stationary data streams without forgetting previously learned knowledge. Mixture models, which add new components when certain conditions are met, have shown promising results in TFCL tasks. However, such approaches do not make use of the knowledge already accumulated for positive knowledge transfer. In this paper, we develop a new model, namely the Online Recursive Variational Autoencoder (ORVAE). ORVAE utilizes the prior knowledge by selectively incorporating the newly learnt information, by adding new components, according to the knowledge already known from the past learnt data. We introduce a new attention mechanism to regularize the structural latent space in which the most important information is reused while the information that interferes with novel samples is inactivated. The proposed attention mechanism can maximize the benefit from the forward transfer for learning novel information without forgetting previously learnt knowledge. We perform several experiments which show that ORVAE achieves state-of-the-art results under TFCL.

AB - Task Free Continual Learning (TFCL) aims to capture novel concepts from non-stationary data streams without forgetting previously learned knowledge. Mixture models, which add new components when certain conditions are met, have shown promising results in TFCL tasks. However, such approaches do not make use of the knowledge already accumulated for positive knowledge transfer. In this paper, we develop a new model, namely the Online Recursive Variational Autoencoder (ORVAE). ORVAE utilizes the prior knowledge by selectively incorporating the newly learnt information, by adding new components, according to the knowledge already known from the past learnt data. We introduce a new attention mechanism to regularize the structural latent space in which the most important information is reused while the information that interferes with novel samples is inactivated. The proposed attention mechanism can maximize the benefit from the forward transfer for learning novel information without forgetting previously learnt knowledge. We perform several experiments which show that ORVAE achieves state-of-the-art results under TFCL.

M3 - Conference contribution

VL - 37

BT - AAAI Conference on Artificial Intelligence

PB - AAAI Press

ER -