# Good Initializations of Variational Bayes for Deep Models

arXiv: https://arxiv.org/abs/1810.08083 · HTML: https://mlsanity.com/paper/1810.08083/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Oct 2018.

## Built on by (most cited first, 11 of 11)

- [Variational Latent Branching Model for Off-Policy Evaluation](https://mlsanity.com/paper/2301.12056/) (2023)
- [On permutation symmetries in Bayesian neural network posteriors: a variational perspective](https://mlsanity.com/paper/2310.10171/) (2023)
- [Uncertainty Quantification for Deep Neural Networks: An Empirical Comparison and Usage Guidelines](https://mlsanity.com/paper/2212.07118/) (2022)
- [Safety Monitoring of Machine Learning Perception Functions: a Survey](https://mlsanity.com/paper/2412.06869/) (2024)
- [Evaluating Bayesian deep learning for radio galaxy classification](https://mlsanity.com/paper/2405.18351/) (2024)
- [Decision Theoretic Bootstrapping](https://mlsanity.com/paper/2103.09982/) (2021)
- [DeepBayes -- an estimator for parameter estimation in stochastic nonlinear dynamical models](https://mlsanity.com/paper/2205.02264/) (2022)
- [Variational Inference for Additive Main and Multiplicative Interaction Effects Models](https://mlsanity.com/paper/2207.00011/) (2022)
- [SoftCVI: Contrastive variational inference with self-generated soft labels](https://mlsanity.com/paper/2407.15687/) (2024)
- [Robust scalable initialization for Bayesian variational inference with multi-modal Laplace approximations](https://mlsanity.com/paper/2307.06424/) (2023)
- [Quantifying Uncertainty in Deep Learning Classification with Noise in Discrete Inputs for Risk-Based Decision Making](https://mlsanity.com/paper/2310.06105/) (2023)

## Abstract (arXiv)

Stochastic variational inference is an established way to carry out approximate Bayesian inference for deep models. While there have been effective proposals for good initializations for loss minimization in deep learning, far less attention has been devoted to the issue of initialization of stochastic variational inference. We address this by proposing a novel layer-wise initialization strategy based on Bayesian linear models. The proposed method is extensively validated on regression and classification tasks, including Bayesian DeepNets and ConvNets, showing faster and better convergence compared to alternatives inspired by the literature on initializations for loss minimization.