Deep neural networks (NNs) are known to lack uncertainty estimates and struggle to incorporate new data. We present a method that mitigates these issues by converting NNs from weight space to function space, via a dual parameterization. Importantly, the dual parameterization enables us to formulate a sparse representation that captures information from the entire data set. This offers a compact and principled way of capturing uncertainty and enables us to incorporate new data without retraining whilst retaining predictive performance. We provide proof-of-concept demonstrations with the proposed approach for quantifying uncertainty in supervised learning on UCI benchmark tasks.
|Number of pages
|Published - 2023
|MoE publication type
|Workshop on Duality Principles for Modern Machine Learning - Honolulu, United States
Duration: 29 Jul 2023 → 29 Jul 2023
|Workshop on Duality Principles for Modern Machine Learning
|29/07/2023 → 29/07/2023