# Reducing numerical precision preserves classification accuracy in Mondrian Forests

arXiv: https://arxiv.org/abs/2106.14340 · HTML: https://mlsanity.com/paper/2106.14340/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 22 references that are in the map.
- First posted on arXiv 28 Jun 2021.

## Builds on (most cited first, 4 of 4)

- [A Study of BFLOAT16 for Deep Learning Training](https://mlsanity.com/paper/1905.12322/) (2019)
- [Training Deep Neural Networks with 8-bit Floating Point Numbers](https://mlsanity.com/paper/1812.08011/) (2018)
- [Mondrian Forests: Efficient Online Random Forests](https://mlsanity.com/paper/1406.2673/) (2014)
- [A benchmark of data stream classification for human activity recognition on connected objects](https://mlsanity.com/paper/2008.11880/) (2020)

## Built on by (most cited first, 2 of 2)

- [Inference with Mondrian Random Forests](https://mlsanity.com/paper/2310.09702/) (2023)
- [Fuzzy PyTorch: Rapid Numerical Variability Evaluation for Deep Learning Models](https://mlsanity.com/paper/2605.25991/) (2026)

## Abstract (arXiv)

Mondrian Forests are a powerful data stream classification method, but their large memory footprint makes them ill-suited for low-resource platforms such as connected objects. We explored using reduced-precision floating-point representations to lower memory consumption and evaluated its effect on classification performance. We applied the Mondrian Forest implementation provided by OrpailleCC, a C++ collection of data stream algorithms, to two canonical datasets in human activity recognition: Recofit and Banos et al. Results show that the precision of floating-point values used by tree nodes can be reduced from 64 bits to 8 bits with no significant difference in F1 score. In some cases, reduced precision was shown to improve classification performance, presumably due to its regularization effect. We conclude that numerical precision is a relevant hyperparameter in the Mondrian Forest, and that commonly-used double precision values may not be necessary for optimal performance. Future work will evaluate the generalizability of these findings to other data stream classifiers.