# Boosting Standard Classification Architectures Through a Ranking Regularizer

arXiv: https://arxiv.org/abs/1901.08616 · HTML: https://mlsanity.com/paper/1901.08616/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 24 Jan 2019.

## Built on by (most cited first, 7 of 7)

- [Context-aware Attentional Pooling (CAP) for Fine-grained Visual Classification](https://mlsanity.com/paper/2101.06635/) (2021)
- [Knowledge Evolution in Neural Networks](https://mlsanity.com/paper/2103.05152/) (2021)
- [Action-slot: Visual Action-centric Representations for Multi-label Atomic Activity Recognition in Traffic Scenes](https://mlsanity.com/paper/2311.17948/) (2023)
- [ATARS: An Aerial Traffic Atomic Activity Recognition and Temporal Segmentation Dataset](https://mlsanity.com/paper/2503.18553/) (2025)
- [Deep Similarity Learning for Sports Team Ranking](https://mlsanity.com/paper/2103.13736/) (2021)
- [SVMax: A Feature Embedding Regularizer](https://mlsanity.com/paper/2103.02770/) (2021)
- [Action-Slot: Structured Action-Centric Representation Learning for Multi-Agent Atomic Activity Understanding](https://mlsanity.com/paper/2609.24127/) (2026)

## Abstract (arXiv)

We employ triplet loss as a feature embedding regularizer to boost classification performance. Standard architectures, like ResNet and Inception, are extended to support both losses with minimal hyper-parameter tuning. This promotes generality while fine-tuning pretrained networks. Triplet loss is a powerful surrogate for recently proposed embedding regularizers. Yet, it is avoided due to large batch-size requirement and high computational cost. Through our experiments, we re-assess these assumptions. During inference, our network supports both classification and embedding tasks without any computational overhead. Quantitative evaluation highlights a steady improvement on five fine-grained recognition datasets. Further evaluation on an imbalanced video dataset achieves significant improvement. Triplet loss brings feature embedding characteristics like nearest neighbor to classification models. Code available at urlhttp://bit.ly/2LNYEqL.