On Lifted Inference Using Neural Embeddings

On Lifted Inference Using Neural Embeddings
复制标题

关于使用神经嵌入的提升推理

DOI:
--
复制
发表时间:
2019
期刊:
AAAI Conference on Artificial Intelligence
影响因子:
--
通讯作者:
D. Venugopal
D. Venugopal
中科院分区:
--
文献类型:
--
作者:
Maminur Islam;Somdeb Sarkhel;D. Venugopal

文献摘要

被引文献

相似文献

我们给出了马尔可夫逻辑网络(MLN)的一种稠密表示,称为Obj2Vec,它编码MLN结构中的对称性。识别对称性是提升推理算法的关键挑战,我们利用神经网络的进步来学习对称性,这些对称性很难使用手工制作的特征来指定。具体地说,我们学习了对MLN对象的嵌入,其预测对象的上下文,即,在MLN的公式中与其一起出现的对象,因为公共上下文指示分布中的对称性。重要的是,我们的公式利用了众所周知的跳过语法模型,使我们能够有效地学习嵌入。最后,为了减小地面MLN的大小,我们基于学习到的嵌入对对象进行采样。我们将Obj2Vec与几种推理算法进行了集成,并与其他最先进的方法进行了比较,证明了该方法的可扩展性和准确性。
We present a dense representation for Markov Logic Networks (MLNs) called Obj2Vec that encodes symmetries in the MLN structure. Identifying symmetries is a key challenge for lifted inference algorithms and we leverage advances in neural networks to learn symmetries which are hard to specify using hand-crafted features. Specifically, we learn an embedding for MLN objects that predicts the context of an object, i.e., objects that appear along with it in formulas of the MLN, since common contexts indicate symmetry in the distribution. Importantly, our formulation leverages well-known skip-gram models that allow us to learn the embedding efficiently. Finally, to reduce the size of the ground MLN, we sample objects based on their learned embeddings. We integrate Obj2Vec with several inference algorithms, and show the scalability and accuracy of our approach compared to other state-of-the-art methods.