Multi-Margin Cosine Loss: An Efficient Loss Function for Recommender Systems

The proposed Multi-Margin Cosine Loss (MMCL) efficiently utilizes not only the hardest negative samples but also other non-trivial negative samples, offering a simpler yet effective loss function that outperforms more complex methods, especially in resource-constrained environments.