Wide Mean-Field Variational Bayesian Neural Networks Ignore the Data.
2021
Variational inference enables approximate posterior inference of the highly over-parameterized neural networks that are popular in modern machine learning. Unfortunately, such posteriors are known to exhibit various pathological behaviors. We prove that as the number of hidden units in a single-layer Bayesian neural network tends to infinity, the function-space posterior mean under mean-field variational inference actually converges to zero, completely ignoring the data. This is in contrast to the true posterior, which converges to a Gaussian process. Our work provides insight into the over-regularization of the KL divergence in variational inference.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
12
References
1
Citations
NaN
KQI