Unified Interpretation of Softmax Cross-Entropy and Negative Sampling: With Case Study for Knowledge Graph Embedding
arXiv:2106.07250 · doi:10.18653/v1/2021.acl-long.429
Abstract
In knowledge graph embedding, the theoretical relationship between the softmax cross-entropy and negative sampling loss functions has not been investigated. This makes it difficult to fairly compare the results of the two different loss functions. We attempted to solve this problem by using the Bregman divergence to provide a unified interpretation of the softmax cross-entropy and negative sampling loss functions. Under this interpretation, we can derive theoretical findings for fair comparison. Experimental results on the FB15k-237 and WN18RR datasets show that the theoretical findings are valid in practical settings.
Accepted at ACL-IJCNLP 2021
References in corpus (5)
- RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space
- Regularizing Neural Networks by Penalizing Confident Output Distributions
- PyKEEN 1.0: A Python Library for Training and Evaluating Knowledge Graph Embeddings
- Notes on Noise Contrastive Estimation and Negative Sampling
- Bregman divergence as general framework to estimate unnormalized statistical models