Distilling Distributional Uncertainty from a Gaussian Process

Page view(s)
19
Checked on Mar 02, 2025
Distilling Distributional Uncertainty from a Gaussian Process
Title:
Distilling Distributional Uncertainty from a Gaussian Process
Journal Title:
ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Keywords:
Publication Date:
18 March 2024
Citation:
Wong, J. H. M., & Chen, N. F. (2024, April 14). Distilling Distributional Uncertainty from a Gaussian Process. ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). https://doi.org/10.1109/icassp48485.2024.10448172
Abstract:
A Neural Network (NN) may exhibit overconfidence about wrong hypotheses, especially for Out-Of-Domain (OOD) inputs. A Gaussian process (GP) instead has an explainable distributional uncertainty behaviour, by predicting hypotheses with greater uncertainty for query inputs further from the training data. Previous work has shown that a NN can learn to emulate the behaviour of a GP on in-domain data. This paper expands upon this, by proposing to train a NN student to emulate the GP teacher's distributional uncertainty behaviour on OOD data. This avoids the computational cost of using a GP at run-time, while improving the OOD confidence calibration of a NN. More accurate confidence calibration may better inform how the system should feedback to the user. Experiments on the SEP-28k-E stutter detection dataset suggest that distillation of such knowledge is feasible between these models.
License type:
Publisher Copyright
Funding Info:
This research / project is supported by the NIE - SpeechEval Phase II: SHE4EDU (Speech Highlighter and Evaluation for Education)
Grant Reference no. : EC-2023-061
Description:
© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
ISBN:
10.1109/ICASSP48485.2024.10448172
Files uploaded:

File Size Format Action
icassp-2024-distil-distributional-uncertainty-from-gaussian-process-revision.pdf 411.27 KB PDF Request a copy