# OPAD: An Optimized Policy-based Active Learning Framework for Document Content Analysis

arXiv: https://arxiv.org/abs/2110.02069 · HTML: https://mlsanity.com/paper/2110.02069/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 36 of its 65 references that are in the map.
- First posted on arXiv 1 Oct 2021.

## Builds on (most cited first, 30 of 36)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Deep Bayesian Active Learning with Image Data](https://mlsanity.com/paper/1703.02910/) (2017)
- [LayoutLM: Pre-training of Text and Layout for Document Image Understanding](https://mlsanity.com/paper/1912.13318/) (2019)
- [Bidirectional LSTM-CRF Models for Sequence Tagging](https://mlsanity.com/paper/1508.01991/) (2015)
- [Generalizing from a Few Examples: A Survey on Few-Shot Learning](https://mlsanity.com/paper/1904.05046/) (2019)
- [LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding](https://mlsanity.com/paper/2012.14740/) (2020)
- [A Survey of Unsupervised Deep Domain Adaptation](https://mlsanity.com/paper/1812.02849/) (2018)
- [PubLayNet: largest dataset ever for document layout analysis](https://mlsanity.com/paper/1908.07836/) (2019)
- [Learning Loss for Active Learning](https://mlsanity.com/paper/1905.03677/) (2019)
- [A Survey on Deep Semi-supervised Learning](https://mlsanity.com/paper/2103.00550/) (2021)
- [Evaluation of Deep Convolutional Nets for Document Image Classification and Retrieval](https://mlsanity.com/paper/1502.07058/) (2015)
- [Cost-Effective Active Learning for Deep Image Classification](https://mlsanity.com/paper/1701.03551/) (2017)
- [Adversarial Active Learning for Deep Networks: a Margin Based Approach](https://mlsanity.com/paper/1802.09841/) (2018)
- [DocBank: A Benchmark Dataset for Document Layout Analysis](https://mlsanity.com/paper/2006.01038/) (2020)
- [Deep Active Learning for Named Entity Recognition](https://mlsanity.com/paper/1707.05928/) (2017)
- [Hierarchical Transformers for Long Document Classification](https://mlsanity.com/paper/1910.10781/) (2019)
- [Learning how to Active Learn: A Deep Reinforcement Learning Approach](https://mlsanity.com/paper/1708.02383/) (2017)
- [DocBERT: BERT for Document Classification](https://mlsanity.com/paper/1904.08398/) (2019)
- [Active Learning for Deep Detection Neural Networks](https://mlsanity.com/paper/1911.09168/) (2019)
- [Reinforced active learning for image segmentation](https://mlsanity.com/paper/2002.06583/) (2020)
- [Adversarial Sampling for Active Learning](https://mlsanity.com/paper/1808.06671/) (2018)
- [Active Learning for Deep Object Detection](https://mlsanity.com/paper/1809.09875/) (2018)
- [Joint Extraction of Events and Entities within a Document Context](https://mlsanity.com/paper/1609.03632/) (2016)
- [dhSegment: A generic deep-learning approach for document segmentation](https://mlsanity.com/paper/1804.10371/) (2018)
- [A Two-Stage Method for Text Line Detection in Historical Documents](https://mlsanity.com/paper/1802.03345/) (2018)
- [Multimodal deep networks for text and image-based document classification](https://mlsanity.com/paper/1907.06370/) (2019)

## Abstract (arXiv)

Documents are central to many business systems, and include forms, reports, contracts, invoices or purchase orders. The information in documents is typically in natural language, but can be organized in various layouts and formats. There have been recent spurt of interest in understanding document content with novel deep learning architectures. However, document understanding tasks need dense information annotations, which are costly to scale and generalize. Several active learning techniques have been proposed to reduce the overall budget of annotation while maintaining the performance of the underlying deep learning model. However, most of these techniques work only for classification problems. But content detection is a more complex task, and has been scarcely explored in active learning literature. In this paper, we propose textitOPAD, a novel framework using reinforcement policy for active learning in content detection tasks for documents. The proposed framework learns the acquisition function to decide the samples to be selected while optimizing performance metrics that the tasks typically have. Furthermore, we extend to weak labelling scenarios to further reduce the cost of annotation significantly. We propose novel rewards to account for class imbalance and user feedback in the annotation interface, to improve the active learning method. We show superior performance of the proposed textitOPAD framework for active learning for various tasks related to document understanding like layout parsing, object detection and named entity recognition. Ablation studies for human feedback and class imbalance rewards are presented, along with a comparison of annotation times for different approaches.