Lifted Relational Neural Networks
This addresses the challenge of combining symbolic reasoning with neural learning for relational data, which is incremental as it builds on existing relational and neural approaches.
The paper tackles the problem of integrating relational-logic representations with neural network learning by proposing a lifted architecture where relational rules serve as templates for unfolding neural networks that share weights across examples. Experiments on 78 relational learning benchmarks demonstrate favorable performance of the method.
We propose a method combining relational-logic representations with neural network learning. A general lifted architecture, possibly reflecting some background domain knowledge, is described through relational rules which may be handcrafted or learned. The relational rule-set serves as a template for unfolding possibly deep neural networks whose structures also reflect the structures of given training or testing relational examples. Different networks corresponding to different examples share their weights, which co-evolve during training by stochastic gradient descent algorithm. The framework allows for hierarchical relational modeling constructs and learning of latent relational concepts through shared hidden layers weights corresponding to the rules. Discovery of notable relational concepts and experiments on 78 relational learning benchmarks demonstrate favorable performance of the method.