Twin Neural Network Regression

Abstract

We introduce twin neural network (TNN) regression. This method predicts differences between the target values of two different data points rather than the targets themselves. The solution of a traditional regression problem is then obtained by averaging over an ensemble of all predicted differences between the targets of an unseen data point and all training data points. Whereas ensembles are normally costly to produce, TNN regression intrinsically creates an ensemble of predictions of twice the size of the training set while only training a single neural network. Since ensembles have been shown to be more accurate than single models this property naturally transfers to TNN regression. We show that TNNs are able to compete or yield more accurate predictions for different data sets, compared to other state-of-the-art methods. Furthermore, TNN regression is constrained by self-consistency conditions. We find that the violation of these conditions provides an estimate for the prediction uncertainty.

References in corpus (8)

• Learning with Pseudo-Ensembles
• Learning to Order Things
• Pitfalls of In-Domain Uncertainty Estimation and Ensembling in Deep Learning
• Discovering Symmetry Invariants and Conserved Quantities by Interpreting Siamese Neural Networks
• Minimum energy path calculations with Gaussian process regression
• Siamese Regression Networks with Efficient mid-level Feature Extraction for 3D Object Pose Estimation
• Hydra: Preserving Ensemble Diversity for Model Distillation
• Twin Neural Network Regression is a Semi-Supervised Regression Algorithm

Cited by in corpus (2)

• Twin Neural Network Regression is a Semi-Supervised Regression Algorithm
• Applying the Case Difference Heuristic to Learn Adaptations from Deep Network Features