AAAI Publications, Thirty-First AAAI Conference on Artificial Intelligence

Font Size: 
Label-Free Supervision of Neural Networks with Physics and Domain Knowledge
Russell Stewart, Stefano Ermon

Last modified: 2017-02-13


In many machine learning applications, labeled data is scarce and obtaining more labels is expensive. We introduce a new approach to supervising neural networks by specifying constraints that should hold over the output space, rather than direct examples of input-output pairs. These constraints are derived from prior domain knowledge, e.g., from known laws of physics. We demonstrate the effectiveness of this approach on real world and simulated computer vision tasks. We are able to train a convolutional neural network to detect and track objects without any labeled examples. Our approach can significantly reduce the need for labeled training data, but introduces new challenges for encoding prior knowledge into appropriate loss functions.


computer vision, unsupervised learning, physics

Full Text: PDF