Hierarchical Indian buffet neural networks for Bayesian continual learning

By Samuel Kessler, Vu Nguyen, Stefan Zohren, Stephen J. Roberts. 2019

26 citations recorded by the source corpus.

We place an Indian Buffet process (IBP) prior over the structure of a Bayesian Neural Network (BNN), thus allowing the complexity of the BNN to increase and decrease automatically. We further extend this model such that the prior on the structure of each hidden layer is shared globally across all layers, using a Hierarchical-IBP (H-IBP). We apply this model to the problem of resource allocation in Continual Learning (CL) where new tasks occur and the network requires extra resources. Our model uses online variational inference with reparameterisation of the Bernoulli and Beta distributions, which constitute the IBP and H-IBP priors. As we automatically learn the number of weights in each layer of the BNN, overfitting and underfitting problems are largely overcome. We show empirically that our approach offers a competitive edge over existing methods in CL.

Subfields: Artificial Intelligence

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Continual Learning via Sequential Function-Space Variational Inference (2023, background): “Other related work on parameter-space regularization includes various modifications to VCL (Ahn et al., 2019; Kessler et al., 2019), uncertainty-guided continual learning in Bayesian neural networks (Ebrahimi et al., 2020), and a variation of SI known as asymmetric loss approximation with…”
Same State, Different Task: Continual Reinforcement Learning without Interference (2021, background): “One approach to CL, referred to as regularization approaches regularizes a NN predictor’s weights to ensure that new learning produces weights which are similar to previous tasks [35, 46, 66, 15, 31].”
Variational Auto-Regressive Gaussian Processes for Continual Learning (2020, methodology): “, 2019), targeting adaptive capacity in Bayesian neural networks (Kessler et al., 2019), and using episodic memory (Lopez-Paz & Ranzato, 2017; Rebuffi et al.”
Continual Learning using a Bayesian Nonparametric Dictionary of Weight Factors (2021, methodology): “The results for HIBNN in table 4 were taken from Kessler et al. (2020), which was trained for 200 epochs.”
VariGrow: Variational Architecture Growing for Task-Agnostic Continual Learning based on Bayesian Novelty (2022, background): “Kessler et al. (2019) proposed a hierarchical Indian Buffet process (IBP) to allocate resources when learning new tasks.”