MLAILGApr 4, 2019

Robust Deep Gaussian Processes

arXiv:1904.02303v217 citations
Originality Synthesis-oriented
AI Analysis

This work addresses robustness issues in DGPs for machine learning practitioners, but it is incremental as it builds on existing methods.

The paper tackles the problem of model misspecification and uncertainty quantification in Deep Gaussian Processes by applying Generalized Variational Inference, resulting in enhancements that improve DGPs with clear interpretations and implementation in under 100 lines of Python code.

This report provides an in-depth overview over the implications and novelty Generalized Variational Inference (GVI) (Knoblauch et al., 2019) brings to Deep Gaussian Processes (DGPs) (Damianou & Lawrence, 2013). Specifically, robustness to model misspecification as well as principled alternatives for uncertainty quantification are motivated with an information-geometric view. These modifications have clear interpretations and can be implemented in less than 100 lines of Python code. Most importantly, the corresponding empirical results show that DGPs can greatly benefit from the presented enhancements.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes