Supplementary Material: Personalizing Gesture Recognition Using Hierarchical Bayesian Neural Networks

For all three datasets, on fifteen random 75-25 split of the data, we trained a Hierarchical Bayesian Neural Network for 100 epochs, with and without using local reparameterization. When not using local reparameterization, we approximated the ELBO using 20 Monte Carlo samples whereas when using local reparameterization, we only used 1 sample. We plot the mean logarithm of the ELBO versus the number of training epochs (Figure 1) and observe that the ELBO curves for the model that employs local reparameterization is much higher than the model that doesn’t, suggesting the model can learn a better approximation of its parameters much faster.