Unified Interpretation of Softmax Cross-Entropy and Negative Sampling: With Case Study for Knowledge Graph Embedding
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.
Code (1)
Tasks
Graph EmbeddingKnowledge Graph EmbeddingLink PredictionvalidMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
UniFace: Unified Cross-Entropy Loss for Deep Face Recognition
As a widely used loss function in deep face recognition, the softmax loss cannot guarantee that the minimum positive sample-to-class similarity is larger than the maximum negative sample-to-class similarity. As a res…
Face RecognitionTARSimilarity search generalisation in contrastive learning with InfoNCE loss
Similarity search is a primary application of embedding models trained by contrastive learning. For one of the most popular contrastive learning loss functions, InfoNCE, we show that the population risk with $k$ negative…
Contrastive LearningSimCE: Simplifying Cross-Entropy Loss for Collaborative Filtering
The learning objective is integral to collaborative filtering systems, where the Bayesian Personalized Ranking (BPR) loss is widely used for learning informative backbones. However, BPR often experiences slow convergence…
Collaborative FilteringRecommendation SystemsRevisiting Softmax Masking: Stop Gradient for Enhancing Stability in Replay-based Continual Learning
In replay-based methods for continual learning, replaying input samples in episodic memory has shown its effectiveness in alleviating catastrophic forgetting. However, the potential key factor of cross-entropy loss with …
Continual LearningIncremental LearningOn Expected Accuracy
We empirically investigate the (negative) expected accuracy as an alternative loss function to cross entropy (negative log likelihood) for classification tasks. Coupled with softmax activation, it has small derivatives o…
ClassificationGeneral Classificationimage-classificationImage Classification+1