# Sparsifying Neural Network Connections for Face Recognition

arXiv: https://arxiv.org/abs/1512.01891 · HTML: https://mlsanity.com/paper/1512.01891/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Dec 2015.

## Built on by (most cited first, 15 of 15)

- [SUTD-TrafficQA: A Question Answering Benchmark and an Efficient Network for Video Reasoning over Traffic Events](https://mlsanity.com/paper/2103.15538/) (2021)
- [DeepPayload: Black-box Backdoor Attack on Deep Learning Models through Neural Payload Injection](https://mlsanity.com/paper/2101.06896/) (2021)
- [IdentityDP: Differential Private Identification Protection for Face Images](https://mlsanity.com/paper/2103.01745/) (2021)
- [SphereFace Revived: Unifying Hyperspherical Face Recognition](https://mlsanity.com/paper/2109.05565/) (2021)
- [PatchCensor: Patch Robustness Certification for Transformers via Exhaustive Testing](https://mlsanity.com/paper/2111.10481/) (2021)
- [AdaSplit: Adaptive Trade-offs for Resource-constrained Distributed Deep Learning](https://mlsanity.com/paper/2112.01637/) (2021)
- [Efficient Sparse Artificial Neural Networks](https://mlsanity.com/paper/2103.07674/) (2021)
- [Learnable Adaptive Cosine Estimator (LACE) for Image Classification](https://mlsanity.com/paper/2110.05324/) (2021)
- [Sparse Super-Regular Networks](https://mlsanity.com/paper/2201.01363/) (2022)
- [End-to-End Sensitivity-Based Filter Pruning](https://mlsanity.com/paper/2204.07412/) (2022)
- [SubFace: Learning with Softmax Approximation for Face Recognition](https://mlsanity.com/paper/2208.11483/) (2022)
- [A Survey on Face Recognition Systems](https://mlsanity.com/paper/2201.02991/) (2022)
- [Pruning by Active Attention Manipulation](https://mlsanity.com/paper/2210.11114/) (2022)
- [Graph Enabled Cross-Domain Knowledge Transfer](https://mlsanity.com/paper/2304.03452/) (2023)
- [Neural Network Pruning as Spectrum Preserving Process](https://mlsanity.com/paper/2307.08982/) (2023)

## Abstract (arXiv)

This paper proposes to learn high-performance deep ConvNets with sparse neural connections, referred to as sparse ConvNets, for face recognition. The sparse ConvNets are learned in an iterative way, each time one additional layer is sparsified and the entire model is re-trained given the initial weights learned in previous iterations. One important finding is that directly training the sparse ConvNet from scratch failed to find good solutions for face recognition, while using a previously learned denser model to properly initialize a sparser model is critical to continue learning effective features for face recognition. This paper also proposes a new neural correlation-based weight selection criterion and empirically verifies its effectiveness in selecting informative connections from previously learned models in each iteration. When taking a moderately sparse structure (26%-76% of weights in the dense model), the proposed sparse ConvNet model significantly improves the face recognition performance of the previous state-of-the-art DeepID2+ models given the same training data, while it keeps the performance of the baseline model with only 12% of the original parameters.