mlsanityData to 7 Oct 2026

Privacy-preserving Online AutoML for Domain-Specific Face Detection

arXiv 2203.08399 ↗First posted 16 Mar 2022cs.CVcited by 4

Not on the map. Why?

Abstract (arXiv)

Despite the impressive progress of general face detection, the tuning of hyper-parameters and architectures is still critical for the performance of a domain-specific face detector. Though existing AutoML works can speedup such process, they either require tuning from scratch for a new scenario or do not consider data privacy. To scale up, we derive a new AutoML setting from a platform perspective. In such setting, new datasets sequentially arrive at the platform, where an architecture and hyper-parameter configuration is recommended to train the optimal face detector for each dataset. This, however, brings two major challenges: (1) how to predict the best configuration for any given dataset without touching their raw images due to the privacy concern? and (2) how to continuously improve the AutoML algorithm from previous tasks and offer a better warm-up for future ones? We introduce "HyperFD", a new privacy-preserving online AutoML framework for face detection. At its core part, a novel meta-feature representation of a dataset as well as its learning paradigm is proposed. Thanks to HyperFD, each local task (client) is able to effectively leverage the learning "experience" of previous tasks without uploading raw images to the platform; meanwhile, the meta-feature extractor is continuously learned to better trade off the bias and variance. Extensive experiments demonstrate the effectiveness and efficiency of our design.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 52 of its 90 references that are in the map. First posted on arXiv 16 Mar 2022.

Builds on

52 of 90 references.

References · 52

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
  4. 2020End-to-End Object Detection with Transformerscited by 7,384
  5. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
  6. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
  7. 2016Overcoming catastrophic forgetting in neural networkscited by 4,718
  8. 2018How Powerful are Graph Neural Networks?cited by 4,355
  9. 2016Learning without Forgettingcited by 2,465
  10. 2016iCaRL: Incremental Classifier and Representation Learningcited by 2,313
  11. 2019MMDetection: Open MMLab Detection Toolbox and Benchmarkcited by 1,379
  12. 2017Continual Learning with Deep Generative Replaycited by 1,152
38 more
  1. 2016Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Networkscited by 895
  2. 2017Lifelong Learning with Dynamically Expandable Networkscited by 689
  3. 2018ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardwarecited by 666
  4. 2019Three scenarios for continual learningcited by 610
  5. 2018BOHB: Robust and Efficient Hyperparameter Optimization at Scalecited by 466
  6. 2018Two at Once: Enhancing Learning and Generalization Capacities via IBN-Netcited by 361
  7. 2015WIDER FACE: A Face Detection Benchmarkcited by 297
  8. 2019Task2Vec: Task Embedding for Meta-Learningcited by 197
  9. 2018Reinforced Continual Learningcited by 186
  10. 2019RetinaFace: Single-stage Dense Face Localisation in the Wildcited by 164
  11. 2016Training Deep Networks for Facial Expression Recognition with Crowd-Sourced Label Distributioncited by 158
  12. 2020Federated Continual Learning with Weighted Inter-client Transfercited by 158
  13. 2019Scalable and Order-robust Continual Learning with Additive Parameter Decompositioncited by 125
  14. 2019Neural Predictor for Neural Architecture Searchcited by 116
  15. 2020BRP-NAS: Prediction-based NAS using GCNscited by 109
  16. 2019DetNAS: Backbone Search for Object Detectioncited by 80
  17. 2019On Network Design Spaces for Visual Recognitioncited by 68
  18. 2020NAS-Bench-NLP: Neural Architecture Search Benchmark for Natural Language Processingcited by 60
  19. 2017FaceBoxes: A CPU Real-time Face Detector with High Accuracycited by 44
  20. 2017SSH: Single Stage Headless Face Detectorcited by 40
  21. 2019Dataset2Vec: Learning Dataset Meta-Featurescited by 37
  22. 2020FBNetV3: Joint Architecture-Recipe Search using Predictor Pretrainingcited by 35
  23. 2017Probabilistic Matrix Factorization for Automated Machine Learningcited by 31
  24. 2018OBOE: Collaborative Filtering for AutoML Model Selectioncited by 31
  25. 2017Understanding and Comparing Deep Neural Networks for Age and Gender Classificationcited by 17
  26. 2018Pushing the Limits of Unconstrained Face Detection: a Challenge Dataset and Baseline Resultscited by 16
  27. 2020Fast Neural Network Adaptation via Parameter Remapping and Architecture Searchcited by 16
  28. 2021FedSpeech: Federated Text-to-Speech with Continual Learningcited by 16
  29. 2018Transfer Learning with Neural AutoMLcited by 15
  30. 2019CenterFace: Joint Face Detection and Alignment Using Face as Pointcited by 13
  31. 2018Global Weighted Average Pooling Bridges Pixel-level Localization and Image-level Classificationcited by 12
  32. 2020HyperSTAR: Task-Aware Hyperparameters for Deep Networkscited by 9
  33. 2020AutoHAS: Efficient Hyperparameter and Architecture Searchcited by 8
  34. 2021CRFace: Confidence Ranker for Model-Agnostic Face Detection Refinementcited by 6
  35. 2021NASOA: Towards Faster Task-oriented Online Fine-tuning with a Zoo of Modelscited by 6
  36. 2014A Fast and Accurate Unconstrained Face Detectorcited by 5
  37. 2019Datasets for Face and Object Detection in Fisheye Imagescited by 5
  38. 2019FDDB-360: Face Detection in 360-degree Fisheye Imagescited by 4

50 most cited of 52 All 52 (JSON)

Built on by

Cited by year

Citing papers per year: 2023: 3; 2024: 1.

4 papers.

  1. 2023MLCopilot: Unleashing the Power of Large Language Models in Solving Machine Learning Taskscited by 66
  2. 2023Training Latency Minimization for Model-Splitting Allowed Federated Edge Learningcited by 3
  3. 2023Online Evolutionary Neural Architecture Search for Multivariate Non-Stationary Time Series Forecastingcited by 2
  4. 2024Utilizing Data Fingerprints for Privacy-Preserving Algorithm Selection in Time Series Classification: Performance and Uncertainty Estimation on Unseen Datasetscited by 0