# Towards Robust Deep Active Learning for Scientific Computing

arXiv: https://arxiv.org/abs/2201.12632 · HTML: https://mlsanity.com/paper/2201.12632/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 43 references that are in the map.
- First posted on arXiv 29 Jan 2022.

## Builds on (most cited first, 13 of 13)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Bounds](https://mlsanity.com/paper/1906.03671/) (2019)
- [A Survey of Deep Active Learning](https://mlsanity.com/paper/2009.00236/) (2020)
- [Tackling Climate Change with Machine Learning](https://mlsanity.com/paper/1906.05433/) (2019)
- [Less is more: sampling chemical space with active learning](https://mlsanity.com/paper/1801.09319/) (2018)
- [Efficient Active Learning for Image Classification and Segmentation using a Sample Selection and Conditional Generative Adversarial Network](https://mlsanity.com/paper/1806.05473/) (2018)
- [Unifying machine learning and quantum chemistry -- a deep neural network for molecular wavefunctions](https://mlsanity.com/paper/1906.10033/) (2019)
- [Generative Adversarial Active Learning](https://mlsanity.com/paper/1702.07956/) (2017)
- [Simulation Intelligence: Towards a New Generation of Scientific Methods](https://mlsanity.com/paper/2112.03235/) (2021)
- [Adversarial Sampling for Active Learning](https://mlsanity.com/paper/1808.06671/) (2018)
- [Benchmarking deep inverse models over time, and the neural-adjoint method](https://mlsanity.com/paper/2009.12919/) (2020)
- [Dropout-based Active Learning for Regression](https://mlsanity.com/paper/1806.09856/) (2018)
- [Modeling nanoconfinement effects using active learning](https://mlsanity.com/paper/2005.02587/) (2020)

## Abstract (arXiv)

Deep learning (DL) is revolutionizing the scientific computing community. To reduce the data gap, active learning has been identified as a promising solution for DL in the scientific computing community. However, the deep active learning (DAL) literature is dominated by image classification problems and pool-based methods. Here we investigate the robustness of pool-based DAL methods for scientific computing problems (dominated by regression) where DNNs are increasingly used. We show that modern pool-based DAL methods all share an untunable hyperparameter, termed the pool ratio, denoted γ, which is often assumed to be known apriori in the literature. We evaluate the performance of five state-of-the-art DAL methods on six benchmark problems if we assume γ is textitnot known - a more realistic assumption for scientific computing problems. Our results indicate that this reduces the performance of modern DAL methods and that they sometimes can even perform worse than random sampling, creating significant uncertainty when used in real-world settings. To overcome this limitation we propose, to our knowledge, the first query synthesis DAL method for regression, termed NA-QBC. NA-QBC removes the sensitive γ hyperparameter and we find that, on average, it outperforms the other DAL methods on our benchmark problems. Crucially, NA-QBC always outperforms random sampling, providing more robust performance benefits.