Completed
Multivariate Heteroscedastic Approach
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Machine Learning Approaches for Predicting Seismic, Acoustic and Atmospheric Data
Automatically move to the next video in the Classroom when playback concludes
- 1 Machine Learning Approaches for Atmospheric and Material Fracture Applications and their Uncertainty Quantification
- 2 My background
- 3 Why Machine Learning (ML) Approaches?
- 4 Why Uncertainty Quantification (UQ) in ML?
- 5 Outline for this presentation
- 6 Problem of Interest and Motivation
- 7 Description of the Data
- 8 Overview of K-Nearest Neighbors Approach
- 9 KNN Graphical Examples
- 10 Our Enhanced KNN-based Approach
- 11 The KNN-based Prediction
- 12 Performance Metrics Options the number of neighbors used
- 13 Figure of Merit in the Space (FMS)
- 14 Normalized Root Mean Squared Error (NRMSE)
- 15 Fraction of Data (FAC2)
- 16 Fractional Bias
- 17 Coefficient of Determination, Slope and Intercept
- 18 Summary: KNN Approach
- 19 Summary: Performance Metrics
- 20 Uncertainty Bounds for Multivariate Machine Learning Predictions on High-Strain Brittle Fracture [1]
- 21 Multivariate Neural Networks Model
- 22 Heteroscedastic Training Loss Function
- 23 Multivariate Heteroscedastic Approach
