Lifelong Mixture of Variational Autoencoders
In this paper, we propose an end-to-end lifelong learning mixture of experts. Each expert is implemented by a Variational Autoencoder (VAE). The experts in the mixture system are jointly trained by maximizing a mixture of individual component evidence lower bounds (MELBO) on the log-likelihood of the given training samples. The mixing coefficients in the mixture, control the contributions of each expert in the goal representation. These are sampled from a Dirichlet distribution whose parameters are determined through non-parametric estimation during lifelong learning. The model can learn new tasks fast when these are similar to those previously learnt. The proposed Lifelong mixture of VAE (L-MVAE) expands its architecture with new components when learning a completely new task. After the training, our model can automatically determine the relevant expert to be used when fed with new data samples. This mechanism benefits both the memory efficiency and the required computational cost as only one expert is used during the inference. The L-MVAE inference model is able to perform interpolation in the joint latent space across the data domains associated with different tasks and is shown to be efficient for disentangled learning representation.
Code (1)
Tasks
Lifelong learningMixture-of-ExpertsSimilar Papers 제목 키워드 기반
Stochastic Backpropagation through Mixture Density Distributions
The ability to backpropagate stochastic gradients through continuous latent distributions has been crucial to the emergence of variational autoencoders and stochastic gradient variational Bayes. The key ingredient is an …
Variational InferenceImproving Variational Autoencoder Estimation from Incomplete Data with Mixture Variational Families
We consider the task of estimating variational autoencoders (VAEs) when the training data is incomplete. We show that missing data increases the complexity of the model's posterior distribution over the latent variables …
ImputationLifelong Generative Modelling Using Dynamic Expansion Graph Model
Variational Autoencoders (VAEs) suffer from degenerated performance, when learning several successive tasks. This is caused by catastrophic forgetting. In order to address the knowledge loss, VAEs are using either Genera…
Lifelong learningmodelRecursive Inference for Variational Autoencoders
Inference networks of traditional Variational Autoencoders (VAEs) are typically amortized, resulting in relatively inaccurate posterior approximation compared to instance-wise variational optimization. Recent semi-amorti…
Variational InferenceMixture of Dynamical Variational Autoencoders for Multi-Source Trajectory Modeling and Separation
In this paper, we propose a latent-variable generative model called mixture of dynamical variational autoencoders (MixDVAE) to model the dynamics of a system composed of multiple moving sources. A DVAE model is pre-train…
Audio Source SeparationMulti-Object TrackingObject TrackingTrajectory Modeling