Classification is a Strong Baseline for Deep Metric Learning
By Andrew Zhai, Hao-Yu Wu. 2018
234 citations recorded by the source corpus.
Subfields: Representation & Self-Supervised Learning, Evaluation & Benchmarks
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Representation & Self-Supervised Learning
- Quoc V. Le
- Andrew Zisserman
- Ludwig Schmidt
- Björn Ommer
- Jitendra Malik
- Arsha Nagrani
- Hieu Pham
- Graham W. Taylor
In Evaluation & Benchmarks
- Jitendra Malik
- Andrew Zisserman
- Quoc V. Le
- Arsha Nagrani
- Björn Ommer
- Serge J. Belongie
- Ludwig Schmidt
- Ser-Nam Lim
Citations in the corpus
- Frozen in Time: A Joint Video and Image Encoder for End-to-End Retrieval (2021, methodology): “We employ [68] in a retrieval setting, where matching text-video pairs in the batch are treated as positives, and all other pairwise combinations in the batch are treated as negatives.”
- Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision (2021, methodology): “In VTAB eval, we follow a hyper-parameter sweep as shown in the Appendix I in (Zhai et al., 2019) with 50 trials for each task.”
- ABO: Dataset and Benchmarks for Real-World 3D Object Understanding (2021, methodology): “Method To evaluate their performance on our multi-view retrieval benchmark, we use the available PyTorch implementations of three recent works that are state-of-theart for their respective approach to deep metric learning: NormSoftmax, (classification-based), ProxyNCA++ (proxy-based) and CGD (triple”
- Characterizing Generalization under Out-Of-Distribution Shifts in Deep Metric Learning (2021, background): “More recent work has tackled sampling complexity through the usage of proxy-representations utilized as sample stand-ins during training, following a NCA [16] objective [41, 26, 64], leveraging softmax-style training through class-proxies [8, 73] or simulating intraclass structures [46].”
- A Metric Learning Reality Check (2020, background): “The most straightforward case is the normalized softmax loss [20,47], which is identical to cross entropy, but with the columns of the weight matrix L2 normalized.”
- ProxyNCA++: Revisiting and Revitalizing Proxy Neighborhood Component Analysis (2020, methodology): “0 - I3 512 NormSoftMax [36] 61.”