# A Framework and Benchmark for Deep Batch Active Learning for Regression

**Authors:** David Holzmüller, Viktor Zaverkin, Johannes Kästner, Ingo Steinwart  
**arXiv:** [2203.09410](https://arxiv.org/abs/2203.09410)  
**Date:** 2022-03-17  
**Summary page:** https://www.summarizepaper.com/en/paper/2203.09410v1/

## Key points

- Framework for constructing Batch Mode Deep Active Learning (BMDAL) algorithms using base kernels, kernel transformations, and selection methods
- Replacing last-layer kernel with sketched Neural Tangent Kernel improves accuracy without significant increases in runtime or memory usage
- Novel LCMD selection method achieves state-of-the-art results in RMSE and MAE on benchmark
- BMDAL methods are scalable and compatible with various network architectures and training methods without modifications
- Limitations include benchmark not covering all potential application scenarios for BMDAL methods
- Study contributes to advancing active learning for regression by introducing new components and demonstrating effectiveness through evaluations on real-world datasets
- Open-source code provided for reproducibility of results and further exploration of BMDAL methods

## Summary

The paper presents a framework for constructing Batch Mode Deep Active Learning (BMDAL) algorithms using base kernels, kernel transformations, and selection methods. The study evaluates various combinations of these components on a benchmark comprising 15 large tabular regression datasets. Results show that replacing the last-layer kernel with a sketched Neural Tangent Kernel leads to improved accuracy without significant increases in runtime or memory usage. Additionally, the novel LCMD selection method achieves state-of-the-art results in terms of Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) on the benchmark. While the BMDAL methods presented in the framework are beneficial for practitioners utilizing neural networks for regression tasks due to their scalability and compatibility with various network architectures and training methods without requiring modifications, there are limitations. The benchmark may not cover all potential application scenarios for the considered BMDAL methods. Overall, the study contributes to advancing the field of active learning for regression by introducing new components and demonstrating their effectiveness through comprehensive evaluations on real-world datasets. The open-source code provided allows for reproducibility of results and further exploration of BMDAL methods in practical applications.