Lifelong generative modeling

Ramapuram, Jason; Gregorova, Magda; Kalousis, Alexandros

doi:10.1016/j.neucom.2020.02.115

Cited by 58 publications

(80 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, both models from [64,67] lack an image inference procedure and Lifelong GAN would need to load all previously learnt data for the generation task. Approaches employing both generative and inference mechanisms are based on the VAE framework [1,50]. However, these approaches have degenerating performance when learning high-dimensional data, due to lacking a powerful generator.…”

Section: Related Workmentioning

confidence: 99%

“…For comparison we consider LGM [50] and VAEGAN [43], which is one of the best known hybrid models enabled with an inference mechanism. We implement VAEGAN using GRM in order to prevent forgetting.…”

Section: Lifelong Unsupervised Learningmentioning

confidence: 99%

“…Figs. 5-a and 5-b show the IS and FID resuls, respectively, where a lower FID and a higher IS indicate better quality images and where we compare with four lifelong learning approaches : LGAN [60], LifelongGAN [67], VAEGAN [43] and LGM [50]. LifelongGAN [67] requires to load the previously learnt real data samples in order to prevent forgetting them when it is applied in general generation tasks (no conditional images are available).…”

Section: Lifelong Unsupervised Learningmentioning

confidence: 99%

See 2 more Smart Citations

Learning Latent Representations Across Multiple Data Domains Using Lifelong VAEGAN

Borş

2020

Computer Vision – ECCV 2020

View full text Add to dashboard Cite

The problem of catastrophic forgetting occurs in deep learning models trained on multiple databases in a sequential manner. Recently, generative replay mechanisms (GRM), have been proposed to reproduce previously learned knowledge aiming to reduce the forgetting. However, such approaches lack an appropriate inference model and therefore can not provide latent representations of data. In this paper, we propose a novel lifelong learning approach, namely the Lifelong VAEGAN (L-VAEGAN), which not only induces a powerful generative replay network but also learns meaningful latent representations, benefiting representation learning. L-VAEGAN can allow to automatically embed the information associated with different domains into several clusters in the latent space, while also capturing semantically meaningful shared latent variables, across different data domains. The proposed model supports many downstream tasks that traditional generative replay methods can not, including interpolation and inference across different data domains.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Lifelong Unsupervised Learningmentioning

confidence: 99%

Section: Lifelong Unsupervised Learningmentioning

confidence: 99%

See 1 more Smart Citation

Learning Latent Representations Across Multiple Data Domains Using Lifelong VAEGAN

Borş

2020

Computer Vision – ECCV 2020

View full text Add to dashboard Cite

show abstract

“…These weights are modified separately during the training (like an annealing procedure) forcing the model to encode information both in the discrete and continuous variables. Moreover, the same model is also used under the setting of continual learning [ 13 ], where a mutual information regularizer is added in order to overcome this issue.…”

Section: Vae With Continuous and Discrete Componentsmentioning

confidence: 99%

“…Unlike in the standard VAE, we can sample data from specific mixture components at will. This is particularly critical if the generative power of VAEs shall be used in conjunction with methods requiring the identification of the distributional components, such as in continual learning [ 13 , 14 ].…”

Section: Introductionmentioning

confidence: 99%

Data-Dependent Conditional Priors for Unsupervised Learning of Multimodal Data

Lavda

Gregorova

Kalousis

2020

Entropy

Self Cite

View full text Add to dashboard Cite

One of the major shortcomings of variational autoencoders is the inability to produce generations from the individual modalities of data originating from mixture distributions. This is primarily due to the use of a simple isotropic Gaussian as the prior for the latent code in the ancestral sampling procedure for data generations. In this paper, we propose a novel formulation of variational autoencoders, conditional prior VAE (CP-VAE), with a two-level generative process for the observed data where continuous z and a discrete c variables are introduced in addition to the observed variables x. By learning data-dependent conditional priors, the new variational objective naturally encourages a better match between the posterior and prior conditionals, and the learning of the latent categories encoding the major source of variation of the original data in an unsupervised manner. Through sampling continuous latent code from the data-dependent conditional priors, we are able to generate new samples from the individual mixture components corresponding, to the multimodal structure over the original data. Moreover, we unify and analyse our objective under different independence assumptions for the joint distribution of the continuous and discrete latent variables. We provide an empirical evaluation on one synthetic dataset and three image datasets, FashionMNIST, MNIST, and Omniglot, illustrating the generative performance of our new model comparing to multiple baselines.

show abstract

The role of lifelong machine learning in bridging the gap between human and machine learning: A scientometric analysis

Abulaish,

Wasi,

Sharma

2024

WIREs Data Min & Knowl

View full text Add to dashboard Cite

Due to advancements in data collection, storage, and processing techniques, machine learning has become a thriving and dominant paradigm. However, one of its main shortcomings is that the classical machine learning paradigm acts in isolation without utilizing the knowledge gained through learning from related tasks in the past. To circumvent this, the concept of Lifelong Machine Learning (LML) has been proposed, with the goal of mimicking how humans learn and acquire cognition. Human learning research has revealed that the brain connects previously learned information while learning new information from a single or small number of examples. Similarly, an LML system continually learns by storing and applying acquired information. Starting with an analysis of how the human brain learns, this paper shows that the LML framework shares a functional structure with the brain when it comes to solving new problems using previously learned information. It also provides a description of the LML framework, emphasizing its similarities to human brain learning. It also provides citation graph generation and scientometric analysis algorithms for the LML literatures, including information about the datasets and evaluation metrics that have been used in the empirical evaluation of LML systems. Finally, it presents outstanding issues and possible future research directions in the field of LML.This article is categorized under: Technologies > Machine Learning

show abstract

Lifelong generative modeling

Cited by 58 publications

References 30 publications

Learning Latent Representations Across Multiple Data Domains Using Lifelong VAEGAN

Learning Latent Representations Across Multiple Data Domains Using Lifelong VAEGAN

Data-Dependent Conditional Priors for Unsupervised Learning of Multimodal Data

The role of lifelong machine learning in bridging the gap between human and machine learning: A scientometric analysis

Contact Info

Product

Resources

About