Learning Distributed Word Representations for Natural Logic Reasoning
This addresses the challenge of integrating logical semantics into distributed representations for natural language processing, though it appears incremental as it builds on existing neural methods.
The paper tackled the problem of training distributed word representations to support natural logic reasoning, using neural network and neural tensor network models, and found positive results on simulated data and the WordNet noun graph.
Natural logic offers a powerful relational conception of meaning that is a natural counterpart to distributed semantic representations, which have proven valuable in a wide range of sophisticated language tasks. However, it remains an open question whether it is possible to train distributed representations to support the rich, diverse logical reasoning captured by natural logic. We address this question using two neural network-based models for learning embeddings: plain neural networks and neural tensor networks. Our experiments evaluate the models' ability to learn the basic algebra of natural logic relations from simulated data and from the WordNet noun graph. The overall positive results are promising for the future of learned distributed representations in the applied modeling of logical semantics.