A Simple Approach to Improve Single-Model Deep Uncertainty via Distance-Awareness

05/01/2022
by   Jeremiah Zhe Liu, et al.
54

Accurate uncertainty quantification is a major challenge in deep learning, as neural networks can make overconfident errors and assign high confidence predictions to out-of-distribution (OOD) inputs. The most popular approaches to estimate predictive uncertainty in deep learning are methods that combine predictions from multiple neural networks, such as Bayesian neural networks (BNNs) and deep ensembles. However their practicality in real-time, industrial-scale applications are limited due to the high memory and computational cost. Furthermore, ensembles and BNNs do not necessarily fix all the issues with the underlying member networks. In this work, we study principled approaches to improve uncertainty property of a single network, based on a single, deterministic representation. By formalizing the uncertainty quantification as a minimax learning problem, we first identify distance awareness, i.e., the model's ability to quantify the distance of a testing example from the training data, as a necessary condition for a DNN to achieve high-quality (i.e., minimax optimal) uncertainty estimation. We then propose Spectral-normalized Neural Gaussian Process (SNGP), a simple method that improves the distance-awareness ability of modern DNNs with two simple changes: (1) applying spectral normalization to hidden weights to enforce bi-Lipschitz smoothness in representations and (2) replacing the last output layer with a Gaussian process layer. On a suite of vision and language understanding benchmarks, SNGP outperforms other single-model approaches in prediction, calibration and out-of-domain detection. Furthermore, SNGP provides complementary benefits to popular techniques such as deep ensembles and data augmentation, making it a simple and scalable building block for probabilistic deep learning. Code is open-sourced at https://github.com/google/uncertainty-baselines

READ FULL TEXT

page 3

page 14

page 19

research
06/17/2020

Simple and Principled Uncertainty Estimation with Deterministic Deep Learning via Distance Awareness

Bayesian neural networks (BNN) and deep ensembles are principled approac...
research
07/29/2020

Hybrid Deep Learning Gaussian Process for Diabetic Retinopathy Diagnosis and Uncertainty Quantification

Diabetic Retinopathy (DR) is one of the microvascular complications of D...
research
07/12/2023

A Bayesian approach to quantifying uncertainties and improving generalizability in traffic prediction models

Deep-learning models for traffic data prediction can have superior perfo...
research
04/08/2023

Deep Anti-Regularized Ensembles provide reliable out-of-distribution uncertainty quantification

We consider the problem of uncertainty quantification in high dimensiona...
research
02/22/2021

Improving Deterministic Uncertainty Estimation in Deep Learning for Classification and Regression

We propose a new model that estimates uncertainty in a single forward pa...
research
07/05/2023

Distance Preserving Machine Learning for Uncertainty Aware Accelerator Capacitance Predictions

Providing accurate uncertainty estimations is essential for producing re...
research
07/19/2021

Epistemic Neural Networks

We introduce the epistemic neural network (ENN) as an interface for unce...

Please sign up or login with your details

Forgot password? Click here to reset