Abstract
Deep Learning models’ performance strongly correlate with availability of annotated data; however, massive data labeling is laborious, expensive, and error-prone when performed by human experts. Active Learning (AL) effectively handles this challenge by selecting the uncertain samples from unlabeled data collection, but the existing AL approaches involve repetitive human feedback for labeling uncertain samples, thus rendering these techniques infeasible to be deployed in industry related real-world applications. In the proposed Proxy Model based Active Learning technique (PMAL), this issue is addressed by replacing human oracle with a deep learning model, where human expertise is reduced to label only two small subsets of data for training proxy model and initializing the AL loop. In the PMAL technique, firstly, proxy model is trained with a small subset of labeled data, which subsequently acts as an oracle for annotating uncertain samples. Secondly, active model's training, uncertain samples extraction via uncertainty sampling, and annotation through proxy model is carried out until predefined iterations to achieve higher accuracy and labeled data. Finally, the active model is evaluated using testing data to verify the effectiveness of our technique for practical applications. The correct annotations by the proxy model are ensured by employing the potentials of explainable artificial intelligence. Similarly, emerging vision transformer is used as an active model to achieve maximum accuracy. Experimental results reveal that the proposed method outperforms the state-of-the-art in terms of minimum labeled data usage and improves the accuracy with 2.2%, 2.6%, and 1.35% on Caltech-101, Caltech-256, and CIFAR-10 datasets, respectively. Since the proposed technique offers a highly reasonable solution to exploit huge multimedia data, it can be widely used in different evolutionary industrial domains.
- [1] . 2021. Introduction to Big Multimodal Multimedia Data with Deep Analytics. 17, ed: ACM New York, NY, 2021, 1–3.Google Scholar
- [2] . 2018. Redundancy avoidance for big data in data centers: A conventional neural network approach. IEEE Transactions on Network Science and Engineering 7, 1 (2018), 104–114.Google Scholar
Cross Ref
- [3] 2021. Fuzzy logic in surveillance big video data analysis: Comprehensive review, challenges, and research directions. ACM Computing Surveys (CSUR) 54, 3 (2021), 1–33.Google Scholar
Digital Library
- [4] . 2021. Where are they going? Predicting human behaviors in crowded scenes. ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM) 17, 4 (2021), 1–19.Google Scholar
Digital Library
- [5] . 2020. Introduction to the Special Issue on Computational Intelligence for Biomedical Data and Imaging. 16, ed: ACM New York, NY, USA, 2020, 1–4.Google Scholar
- [6] . 2021. eDiaPredict: An ensemble-based framework for diabetes prediction. ACM Transactions on Multimedia Computing Communications and Applications 17, 2s (2021), 1–26.Google Scholar
Digital Library
- [7] . 2021. Mill: Channel attention–based deep multiple instance learning for landslide recognition. ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM) 17, 2s (2021), 1–11.Google Scholar
Digital Library
- [8] . 2022. An effective forest fire detection framework using heterogeneous wireless multimedia sensor networks. ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM) 18, 2 (2022), 1–21.Google Scholar
Digital Library
- [9] 2021. Machine learning-based Mist computing enabled internet of battlefield things. ACM Transactions on Internet Technology (TOIT) 21, 4 (2021), 1–26.Google Scholar
Digital Library
- [10] . 2021. A hybrid deep reinforcement learning for autonomous vehicles smart-platooning. IEEE Transactions on Vehicular Technology (2021).Google Scholar
Cross Ref
- [11] . 2021. QuickLook: Movie summarization using scene-based leading characters with psychological cues fusion. Information Fusion 76 (2021), 24–35.Google Scholar
Digital Library
- [12] . 2022. MMSUM digital twins: A multi-view multi-modality summarization framework for sporting events. ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM) 18, 1 (2022), 1–25.Google Scholar
Digital Library
- [13] 2021. An adaptive neural architecture search design for collaborative edge-cloud computing. IEEE Network 35, 5 (2021), 83–89.Google Scholar
Digital Library
- [14] . 2021. Data engineering, preparation, and labeling for AI 2019. https://www.cognilytica.com/document/report-data-engineering-preparation-and-labeling-for-ai-2019/ (accessed 29/11/2021, 2021).Google Scholar
- [15] I. Grand View Research. 2021. Data collection and labeling market worth $8.22 billion by 2028. https://www.grandviewresearch.com/press-release/global-data-collection-labeling-market (accessed 29/11/2021).Google Scholar
- [16] 2021. A survey of deep active learning. ACM Computing Surveys (CSUR) 54, 9 (2021), 1–40.Google Scholar
Digital Library
- [17] . 2016. Cost-effective active learning for deep image classification. IEEE Transactions on Circuits and Systems for Video Technology 27, 12 (2016), 2591–2600.Google Scholar
Digital Library
- [18] . 2021. Survey on recent active learning methods for deep learning. In Advances in Parallel & Distributed Processing, and Applications. Springer, 609–617.Google Scholar
Cross Ref
- [19] . 2009. Active learning literature survey. 2009.Google Scholar
- [20] . 2009. Active learning methods for remote sensing image classification. IEEE Transactions on Geoscience and Remote Sensing 47, 7 (2009), 2218–2232.Google Scholar
Cross Ref
- [21] . 2020. Data labeling: An empirical investigation into industrial challenges and mitigation strategies. In International Conference on Product-Focused Software Process Improvement. Springer, 202–216.Google Scholar
Digital Library
- [22] . 2017. Active learning for convolutional neural networks: A core-set approach. arXiv preprint arXiv:1708.00489.Google Scholar
- [23] 2020. Minimax active learning. arXiv preprint arXiv:2012.10467.Google Scholar
- [24] . 2016. Budgeted stream-based active learning via adaptive submodular maximization. Advances in Neural Information Processing Systems 29, 2016.Google Scholar
- [25] . 2018. Adversarial active learning for deep networks: A margin based approach. arXiv preprint arXiv:1802.09841.Google Scholar
- [26] . 2019. Bayesian generative active deep learning. In International Conference on Machine Learning. PMLR, 6295–6304.Google Scholar
- [27] . 2010. Self-paced learning for latent variable models. Advances in Neural Information Processing Systems 23, (2010).Google Scholar
- [28] . 2013. Feedback-driven multiclass active learning for data streams. In Proceedings of the 22nd ACM International Conference on Information & Knowledge Management. 1311–1320.Google Scholar
Digital Library
- [29] . 2014. Stream-based active learning for sentiment analysis in the financial domain. Information Sciences 285 (2014), 181–203.Google Scholar
Digital Library
- [30] . 2020. Active learning query strategies for classification, regression, and clustering: A survey. Journal of Computer Science and Technology 35, 4 (2020), 913–945.Google Scholar
Digital Library
- [31] . 2018. Query-by-committee improvement with diversity and density in batch active learning. Information Sciences 454 (2018), 401–418.Google Scholar
Cross Ref
- [32] . 2013. Adaptive active learning for image classification. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 859–866.Google Scholar
Digital Library
- [33] . 2021. Active learning approach using a modified least confidence sampling strategy for named entity recognition. Progress in Artificial Intelligence 10, 2 (2021), 113–128.Google Scholar
Digital Library
- [34] . 2013. A convex optimization framework for active learning. In Proceedings of the IEEE International Conference on Computer Vision. 209–216.Google Scholar
Digital Library
- [35] . 2003. Incorporating diversity in active learning with support vector machines. In Proceedings of the 20th International Conference on Machine Learning (ICML'03). 59–66.Google Scholar
- [36] . 2020. Adversarial sampling for active learning. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 3071–3079.Google Scholar
- [37] . 2015. Multi-class active learning by uncertainty sampling with diversity maximization. International Journal of Computer Vision 113, 2 (2015), 113–127.Google Scholar
Digital Library
- [38] . 2021. Task-aware variational adversarial active learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 8166–8175.Google Scholar
Cross Ref
- [39] . 2021. Self-paced active learning for deep CNNs via effective loss function. Neurocomputing 424 (2021), 1–8.Google Scholar
Cross Ref
- [40] . 2021. MCDAL: Maximum classifier discrepancy for active learning. arXiv preprint arXiv:2107.11049.Google Scholar
- [41] . 2016. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 770–778.Google Scholar
Cross Ref
- [42] . 2007. Pattern Recognition and Neural Networks. Cambridge University Press (2007).Google Scholar
- [43] . 2019. Wider or deeper: Revisiting the ResNet model for visual recognition. Pattern Recognition 90 (2019), 119–133.Google Scholar
Digital Library
- [44] . 2018. A survey on deep transfer learning. In International Conference on Artificial Neural Networks. Springer, 270–279.Google Scholar
Cross Ref
- [45] 2015. ImageNet large scale visual recognition challenge. International Journal of Computer Vision 115, 3 (2015), 211–252.Google Scholar
Digital Library
- [46] 2017. Attention is all you need. In Advances in Neural Information Processing Systems. 5998–6008.Google Scholar
Digital Library
- [47] 2020. An image is worth 16 × 16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929.Google Scholar
- [48] . 2021. Transformers in vision: A survey. arXiv preprint arXiv:2101.01169.Google Scholar
- [49] . Revisiting Uncertainty-based Query Strategies for Active Learning with Transformers.Google Scholar
- [50] . 2009. Learning multiple layers of features from tiny images. (2009).Google Scholar
- [51] . 2006. One-shot learning of object categories. IEEE Transactions on Pattern Analysis and Machine Intelligence 28, 4 (2006), 594–611.Google Scholar
Digital Library
- [52] . 2007. Caltech-256 object category dataset. (2007).Google Scholar
- [53] . 2014. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.Google Scholar
- [54] . 2019. EfficientNet: Rethinking model scaling for convolutional neural networks. Presented at the Proceedings of the 36th International Conference on Machine Learning, Proceedings of Machine Learning Research (2019). [Online]. Available: https://proceedings.mlr.press/v97/tan19a.html.Google Scholar
Index Terms
PMAL: A Proxy Model Active Learning Approach for Vision Based Industrial Applications
Recommendations
Active Learning from Positive and Unlabeled Data
ICDMW '11: Proceedings of the 2011 IEEE 11th International Conference on Data Mining WorkshopsDuring recent years, active learning has evolved into a popular paradigm for utilizing user's feedback to improve accuracy of learning algorithms. Active learning works by selecting the most informative sample among unlabeled data and querying the label ...
Active Learning for kNN Using Instance Impact
AI 2022: Advances in Artificial IntelligenceAbstractLabelling unlabeled data is a time-consuming and expensive process. Labelling initiatives should select samples that are likely to enhance the classification accuracy of the classifier. Several methods can be employed to accomplish this goal. One ...
Graph-Based Active Learning Based on Label Propagation
MDAI '08 Sabadell: Proceedings of the 5th International Conference on Modeling Decisions for Artificial IntelligenceBy only selecting the most informative instances for labeling, active learning could reduce the labeling cost when labeled instances are hard to obtain. Facing the same situation, semi-supervised learning utilize unlabeled instances to strengthen ...






Comments