MLLGApr 5, 2023

Self-Distillation for Gaussian Process Regression and Classification

arXiv:2304.02641v12 citationsh-index: 3
Originality Synthesis-oriented
AI Analysis

This work addresses the problem of applying knowledge distillation to Gaussian Process models for researchers in machine learning, but it appears incremental as it adapts existing distillation concepts to a specific model type.

The authors tackled the problem of extending knowledge distillation to Gaussian Process Regression and Classification by proposing data-centric and distribution-centric approaches, showing that these methods relate to existing techniques like kernel ridge regression self-distillation and data duplication, and they claim to be the first to formulate distillation specifically for Gaussian Process models.

We propose two approaches to extend the notion of knowledge distillation to Gaussian Process Regression (GPR) and Gaussian Process Classification (GPC); data-centric and distribution-centric. The data-centric approach resembles most current distillation techniques for machine learning, and refits a model on deterministic predictions from the teacher, while the distribution-centric approach, re-uses the full probabilistic posterior for the next iteration. By analyzing the properties of these approaches, we show that the data-centric approach for GPR closely relates to known results for self-distillation of kernel ridge regression and that the distribution-centric approach for GPR corresponds to ordinary GPR with a very particular choice of hyperparameters. Furthermore, we demonstrate that the distribution-centric approach for GPC approximately corresponds to data duplication and a particular scaling of the covariance and that the data-centric approach for GPC requires redefining the model from a Binomial likelihood to a continuous Bernoulli likelihood to be well-specified. To the best of our knowledge, our proposed approaches are the first to formulate knowledge distillation specifically for Gaussian Process models.

Code Implementations1 repo
Foundations

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

Your Notes