mlsanityData to 14 Oct 2026

Stream-based Active Learning by Exploiting Temporal Properties in Perception with Temporal Predicted Loss

arXiv 2309.05517 ↗First posted 11 Sep 2023cs.CV cs.LGcited by 6

Not on the map. Why?

Abstract (arXiv)

Active learning (AL) reduces the amount of labeled data needed to train a machine learning model by intelligently choosing which instances to label. Classic pool-based AL requires all data to be present in a datacenter, which can be challenging with the increasing amounts of data needed in deep learning. However, AL on mobile devices and robots, like autonomous cars, can filter the data from perception sensor streams before reaching the datacenter. We exploited the temporal properties for such image streams in our work and proposed the novel temporal predicted loss (TPL) method. To evaluate the stream-based setting properly, we introduced the GTA V streets and the A2D2 streets dataset and made both publicly available. Our experiments showed that our approach significantly improves the diversity of the selection while being an uncertainty-based method. As pool-based approaches are more common in perception applications, we derived a concept for comparing pool-based and stream-based AL, where TPL out-performed state-of-the-art pool- or stream-based approaches for different models. TPL demonstrated a gain of 2.5 precept points (pp) less required data while being significantly faster than pool-based methods.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 22 of its 38 references that are in the map. First posted on arXiv 11 Sep 2023.

Builds on

22 of 38 references.

References · 22

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  4. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  5. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,444
  6. 2018Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentationcited by 3,223
  7. 2019Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Boundscited by 610
  8. 2020A Survey of Deep Active Learningcited by 554
  9. 2019BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learningcited by 432
  10. 2019Learning Loss for Active Learningcited by 336
  11. 2019Variational Adversarial Active Learningcited by 284
  12. 2020A2D2: Audi Autonomous Driving Datasetcited by 190
10 more
  1. 2020Task-Aware Variational Adversarial Active Learningcited by 84
  2. 2020Sequential Graph Convolutional Network for Active Learningcited by 69
  3. 2018Efficient Uncertainty Estimation for Semantic Segmentation in Videoscited by 46
  4. 2019Deep Active Learning for Efficient Training of a LiDAR 3D Object Detectorcited by 45
  5. 2019Temporal Coherence for Active Learning in Videoscited by 24
  6. 2019Submodular Streaming in All its Glory: Tight Approximation, Minimum Memory and Low Adaptive Complexitycited by 14
  7. 2022Stream-based active learning with linear modelscited by 9
  8. 2021Bayesian Active Learning for Sim-to-Real Robotic Perceptioncited by 8
  9. 2021Active Learning for Deep Neural Networks on Edge Devicescited by 6
  10. 2017Episode-Based Active Learning with Bayesian Neural Networkscited by 4

Built on by

Cited by year

Citing papers per year: 2024: 3; 2025: 3.

6 papers.

  1. 2024Uncertainty for Active Learning on Graphscited by 12
  2. 2024A Unified Approach Towards Active Learning and Out-of-Distribution Detectioncited by 7
  3. 2025Joint Out-of-Distribution Filtering and Data Discovery Active Learningcited by 7
  4. 2025Prior2Former -- Evidential Modeling of Mask Transformers for Assumption-Free Open-World Panoptic Segmentationcited by 7
  5. 2024Stream-Based Active Learning for Process Monitoringcited by 0
  6. 2025A Machine Learning Perspective on Automated Driving Corner Casescited by 0