Affine Invariant Ensemble Transform Methods to Improve Predictive Uncertainty in Neural Networks
This work addresses predictive uncertainty in neural networks, which is crucial for reliable AI applications, but it appears incremental as it builds on existing ensemble Kalman filter methods.
The authors tackled the problem of Bayesian inference for logistic regression by proposing two interacting particle systems based on ensemble Kalman filter extensions, proving convergence rates to their mean-field limit, and applied these to improve predictive uncertainty quantification in neural networks.
We consider the problem of performing Bayesian inference for logistic regression using appropriate extensions of the ensemble Kalman filter. Two interacting particle systems are proposed that sample from an approximate posterior and prove quantitative convergence rates of these interacting particle systems to their mean-field limit as the number of particles tends to infinity. Furthermore, we apply these techniques and examine their effectiveness as methods of Bayesian approximation for quantifying predictive uncertainty in neural networks.