NOMU: Neural Optimization-based Model Uncertainty

02/26/2021
by   Jakob Heiss, et al.
7

We introduce a new approach for capturing model uncertainty for neural networks (NNs) in regression, which we call Neural Optimization-based Model Uncertainty (NOMU). The main idea of NOMU is to design a network architecture consisting of two connected sub-networks, one for the model prediction and one for the model uncertainty, and to train it using a carefully designed loss function. With this design, NOMU can provide model uncertainty for any given (previously trained) NN by plugging it into the framework as the sub-network used for model prediction. NOMU is designed to yield uncertainty bounds (UBs) that satisfy four important desiderata regarding model uncertainty, which established methods often do not satisfy. Furthermore, our UBs are themselves representable as a single NN, which leads to computational cost advantages in applications such as Bayesian optimization. We evaluate NOMU experimentally in multiple settings. For regression, we show that NOMU performs as well as or better than established benchmarks. For Bayesian optimization, we show that NOMU outperforms all other benchmarks.

READ FULL TEXT

page 13

page 19

page 20

page 23

page 24

page 25

research
12/08/2021

Calibration Improves Bayesian Optimization

Bayesian optimization is a procedure that allows obtaining the global op...
research
07/12/2019

Differentiable Bayesian Neural Network Inference for Data Streams

While deep neural networks (NNs) do not provide the confidence of its pr...
research
06/03/2019

Quantifying Point-Prediction Uncertainty in Neural Networks via Residual Estimation with an I/O Kernel

Neural Networks (NNs) have been extensively used for a wide spectrum of ...
research
08/26/2020

Uncertainty-Aware Surrogate Model For Oilfield Reservoir Simulation

Deep neural networks have gained increased attention in machine learning...
research
11/26/2020

CYPUR-NN: Crop Yield Prediction Using Regression and Neural Networks

Our recent study using historic data of paddy yield and associated condi...
research
01/22/2020

Bayesian design for minimising uncertainty in spatial processes

Model-based geostatistical design involves the selection of locations to...
research
06/24/2020

Uncertainty in Neural Relational Inference Trajectory Reconstruction

Neural networks used for multi-interaction trajectory reconstruction lac...

Please sign up or login with your details

Forgot password? Click here to reset