skip to primary navigationskip to content

Cambridge Service Alliance

At the forefront of service transformation in the digital era

Studying at Cambridge

 

Webinar - High-Quality Prediction Intervals for Deep Learning

last modified Jun 11, 2018 09:51 PM
11 June 2018 - High-Quality Prediction Intervals for Deep Learning: A Distribution-Free, Ensembled Approach - by Tim Pearce, Mohamed Zaki, Alexandra Brintrup, Andy Neely

In this webinar, Tim discusses the February Alliance paper. Deep neural networks (NNs) have caused great excitement due to the step-changes in performance they have delivered in a variety of applications. However, their appeal in industry can be inhibited by an inability to quantify the uncertainty of their predictions. To take a prognostics example, a typical NN might predict that a machine will fail in 60 days. It is unclear from this point prediction whether the machine should be repaired immediately, or whether it can be run for another 59 days. However, if the NN could output a prediction interval (PI) of 45-65 days with 99% probability, timing of a repair could easily be scheduled. In this paper, we develop a method for doing exactly this - the quantification of uncertainty in deep learning using PIs. We derive a method based on the assumption that high-quality PIs should be as narrow as possible, whilst still capturing a given proportion of data. The method is general, applicable to any data-driven task where a continuous value needs to be predicted, and it is important to know the uncertainty of that prediction. Examples include the forecasting of precipitation, energy load, financial metrics, and traffic volume. The method is tested on ten real-world, open-source datasets. The proposed method is shown to outperform current state-of-the-art uncertainty quantification methods, reducing average PI width by around 10%.

Webinar I Presentation

RSS Feed Latest news

Managing customer experience through a systematic approach to data analytics

Jul 24, 2020

We all know what a positive – or negative – customer experience feels like: a rush of pleasure or teeth-grinding frustration. Firms are acutely aware of the importance of how we feel – and how that translates into long-term loyalty – but feelings are a difficult and complicated thing to measure at scale.

Blog | Has COVID-19 Invented the 10X Supply Chain?

Jul 06, 2020

Surendra Kancherla, Principal Architect - Digital Technologies, at CSA partner, HCL, says that COVID-19 is forcing firms to look afresh at their supply chain capabilities.

Insight | Business model innovation for digital twins in the construction sector

May 19, 2020

At the Cambridge Service Alliance's most recent Community of Interest, Dr Erika Pärn considered the development of digital twins in the construction sector: the challenges it faces and whether android operating systems could provide a useful roadmap for business model innovation.

Generate revenue from digital platforms

May 12, 2020

According to Shep Hyken: "Customers are smarter than ever because they're experiencing great service from certain brands, and those rock-star brands are setting the bar higher for everyone else." In this white paper we explore the 'customer experience paradox' and how firms can use digital platforms to overcome it.

Unlocking the secrets of customer loyalty

May 12, 2020

Customer retention plays a critical role in a firm's long-term sustainability. Until recently, however, firms have tended to rely on one-dimensional survey techniques to measure their customers' loyalty. This white paper shows how machine learning can help overcome these limitations.

View all news