NTR Webinar: Uncertainty Estimation: Can Your Neural Network Provide Confidence For its Predictions?

NTR organizes and hosts scientific webinars on neural networks and invites speakers from all over the world to present their recent work. 

On December 7 Maxim Panov, Skoltech, Moscow, Russia, presented a technical Zoom webinar on Can Your Neural Network Provide Confidence For its Predictions?

About the webinar: 

Neural networks perform very well in almost all machine learning applications. However, neural networks often make very confident predictions for out-of-sample data or data at the boundary between classes. 

In many applications, this is unacceptable, and therefore the ability to assess the degree of confidence in a prediction is extremely important and required.

However, the estimation of uncertainty for neural networks is a non-trivial task, and the existing approaches demonstrated are not very high quality and often require significant computational resources. 

In the presentation, we discussed existing approaches to uncertainty estimation, including model calibration methods, ensemble construction methods, and Bayesian neural networks. 

Special attention was paid to modern numerically efficient approaches based on a single neural network, which do not require the construction of an ensemble and a significant change in the training procedure.

Webinar presentation.

Moderator and contact: NTR CEO Nick Mikhailovsky: nickm@ntrlab.com.

Leave a Reply

Your email address will not be published. Required fields are marked *