Learning Uncertainty with Artificial Neural Networks for Improved Remaining Time Prediction of Business Processes (2021-05-12T00:00:00.000000Z)

TL;DR

This paper is the first to apply Bayesian neural networks' uncertainty estimates themselves to predictive process monitoring and found that they contribute towards more accurate predictions and work quickly.

Abstract

Artificial neural networks will always make a prediction, even when completely uncertain and regardless of the consequences. This obliviousness of uncertainty is a major obstacle towards their adoption in practice. Techniques exist, however, to estimate the two major types of uncertainty: model uncertainty and observation noise in the data. Bayesian neural networks are theoretically well-founded models that can learn the model uncertainty of their predictions. Minor modifications to these models and their loss functions allow learning the observation noise for individual samples as well. This paper is the first to apply these techniques to predictive process monitoring. We found that they contribute towards more accurate predictions and work quickly. However, their main benefit resides with the uncertainty estimates themselves that allow the separation of higher-quality from lower-quality predictions and the building of confidence intervals. This leads to many interesting applications, enables an earlier adoption of prediction systems with smaller datasets and fosters a better cooperation with humans.

Authors

Jochen De Weerdt

4 papers

Hans Weytjens

1 papers

Learning Uncertainty with Artificial Neural Networks for Improved Remaining Time Prediction of Business Processes

TL;DR

Abstract

Authors

References20 items

Process Outcome Prediction: CNN vs. LSTM (with Attention)

Survey and Cross-benchmark Comparison of Remaining Time Prediction Methods in Business Process Monitoring

An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling

LSTM networks for data-aware remaining time prediction of business process instances

Concrete Dropout

What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?

Predictive Business Process Monitoring with LSTM Neural Networks

The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables

A Theoretically Grounded Application of Dropout in Recurrent Neural Networks

Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning

Bayesian Convolutional Neural Networks with Bernoulli Approximate Variational Inference

Data-aware remaining time prediction of business process instances

Auto-Encoding Variational Bayes

Time prediction based on process mining

Cycle Time Prediction: When Will This Case Finally Be Finished?

Long Short-Term Memory

Uncertainty in deep learning: PhD thesis,

Dropout: a simple way to prevent neural networks from overfitting

Bayesian methods for neural networks: theory and applications

Using the heteroscedastic NNs with concrete dropout, we apply MC sampling ( T stochastic forward passes) and average to calculate point estimates (Equation 3)

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names