DL2: Training and Querying Neural Networks with Logic
[edit]
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:19311941, 2019.
Abstract
We present DL2, a system for training and querying neural networks with logical constraints. Using DL2, one can declaratively specify domain knowledge constraints to be enforced during training, as well as pose queries on the model to find inputs that satisfy a set of constraints. DL2 works by translating logical constraints into a loss function with desirable mathematical properties. The loss is then minimized with standard gradientbased methods. We evaluate DL2 by training networks with interesting constraints in unsupervised, semisupervised and supervised settings. Our experimental evaluation demonstrates that DL2 is more expressive than prior approaches combining logic and neural networks, and its loss functions are better suited for optimization. Further, we show that for a number of queries, DL2 can find the desired inputs in seconds (even for large models such as ResNet50 on ImageNet).
Related Material


