Sobolev Training for Neural Networks
Authors:
- Wojciech Marian CzarneckiWCWojciech Marian Czarnecki
- Simon OsinderoSOSimon Osindero
- Max JaderbergMJMax Jaderberg
- Grzegorz ŚwirszczGŚGrzegorz Świrszcz
- Razvan PascanuRPRazvan Pascanu
Preprint
Abstract
At the heart of deep learning we aim to use neural networks as function approximators - training them to produce outputs from inputs in emulation of a ground truth function or data creation process. In many cases we only have access to input-output pairs from the ground truth, however it is becoming more common to have access to derivatives of the target output with respect to the input - for example when the ground truth function is itself a neural network such as in network compression or distillation. Generally these target derivatives are not computed, or are ignored. This paper introduces Sobolev Training for neural networks, which is a method for incorporating these target derivatives in addition the to target values while training. By optimising neural networks to not only approximate the function's outputs but also the function's derivatives we encode additional information about the target function within the parameters of the neural network. Thereby we can improve the quality of our predictors, as well as the data-efficiency and generalization capabilities of our learned function approximation. We provide theoretical justifications for such an approach as well as examples of empirical evidence on three distinct domains: regression on classical optimisation datasets, distilling policies of an agent playing Atari, and on large-scale applications of synthetic gradients. In all three domains the use of Sobolev Training, employing target derivatives in addition to target values, results in models with higher accuracy and stronger generalisation.Keywords
All Related Versions
This publication has 4 references indexed in Scilit:
- A Connection Between Score Matching and Denoising AutoencodersNeural Computation, 2011
- Least Squares Solutions of the HJB Equation With Neural Network Value-Function ApproximatorsIEEE Transactions on Neural Networks, 2007
- On learning the derivatives of an unknown mapping with multilayer feedforward networksNeural Networks, 1992
- Approximation capabilities of multilayer feedforward networksNeural Networks, 1991