Towards Universal Backward-Compatible Representation Learning
Authors: Binjie Zhang, Yixiao Ge, Yantao Shen, Shupeng Su, Fanzi Wu, Chun Yuan, Xuyuan Xu, Yexin Wang, Ying Shan
IJCAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Comprehensive experiments on the large-scale face recognition datasets MS1Mv3 and IJB-C fully demonstrate the effectiveness of our method. |
| Researcher Affiliation | Collaboration | Binjie Zhang1,2 , Yixiao Ge2 , Yantao Shen4 , Shupeng Su2 , Fanzi Wu4 , Chun Yuan1 , Xuyuan Xu3 , Yexin Wang3 , Ying Shan2 1Tsinghua University 2ARC Lab, Tencent PCG 3AI Technology Center of Tencent Video 4AWS/Amazon AI |
| Pseudocode | No | The paper describes its method using textual explanations and mathematical equations (e.g., Equations 1-12) but does not provide structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | Source code is available at https://github.com/Tencent ARC/Open Compatible. |
| Open Datasets | Yes | MS-Celeb-1M (MS1M) [Guo et al., 2016] is a large-scale face recognition training dataset... we adopt MS1Mv3 [Deng et al., 2019] as the training set, which is made up of 5,179,510 training images with 93,431 labels.IJB-C [Maze et al., 2018], a challenging benchmark, is utilized as the open-set evaluation dataset, which has around 1.3 million images. |
| Dataset Splits | No | The paper describes various training data allocations (e.g., "Extended-data", "Open-data") and their sizes in Table 1, but it does not specify a separate "validation" split used during model training to tune hyperparameters or for early stopping. |
| Hardware Specification | Yes | We use 4 NVIDIA V100 GPUs for training. |
| Software Dependencies | No | The paper mentions using "Arc Face loss" and links to "insightface" but does not provide specific version numbers for software dependencies such as Python, PyTorch/TensorFlow, or other libraries. |
| Experiment Setup | Yes | The learning rate is set to 0.1 and decreases 10 times at the 20th, 26th and 32th epoch. The training stops after 35 epochs. The weight decay is set to 10 4 and momentum is 0.9. Batch size is set to 256. The scale factor s and margin m in Eq. 4 are 64, 0.5 following the default setting . In graph-based prototype refinement, we set λ to 0.9, T to 0.05. |