# Getting deep recommenders fit: Bloom embeddings for sparse binary input/output networks

arXiv: https://arxiv.org/abs/1706.03993 · HTML: https://mlsanity.com/paper/1706.03993/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Jun 2017.

## Built on by (most cited first, 14 of 14)

- [FiNER: Financial Numeric Entity Recognition for XBRL Tagging](https://mlsanity.com/paper/2203.06482/) (2022)
- [A Comprehensive Survey on Trustworthy Recommender Systems](https://mlsanity.com/paper/2209.10117/) (2022)
- [MUDES: Multilingual Detection of Offensive Spans](https://mlsanity.com/paper/2102.09665/) (2021)
- [Embedding in Recommender Systems: A Survey](https://mlsanity.com/paper/2310.18608/) (2023)
- [Binary Code based Hash Embedding for Web-scale Applications](https://mlsanity.com/paper/2109.02471/) (2021)
- [Nimble GNN Embedding with Tensor-Train Decomposition](https://mlsanity.com/paper/2206.10581/) (2022)
- [Position-based Hash Embeddings For Scaling Graph Neural Networks](https://mlsanity.com/paper/2109.00101/) (2021)
- [Mem-Rec: Memory Efficient Recommendation System using Alternative Representation](https://mlsanity.com/paper/2305.07205/) (2023)
- [TEI2GO: A Multilingual Approach for Fast Temporal Expression Identification](https://mlsanity.com/paper/2403.16804/) (2024)
- [Probabilistic Hash Embeddings for Online Learning of Categorical Features](https://mlsanity.com/paper/2511.20893/) (2025)
- [Application of the Multi-label Residual Convolutional Neural Network text classifier using Content-Based Routing process](https://mlsanity.com/paper/2110.15801/) (2021)
- [An Ensemble Approach to Acronym Extraction using Transformers](https://mlsanity.com/paper/2201.03026/) (2022)
- [Kamae: Bridging Spark and Keras for Seamless ML Preprocessing](https://mlsanity.com/paper/2507.06021/) (2025)
- [Sparse PPMI Graph Averaging for Random Indexing Embeddings](https://mlsanity.com/paper/2608.05724/) (2026)

## Abstract (arXiv)

Recommendation algorithms that incorporate techniques from deep learning are becoming increasingly popular. Due to the structure of the data coming from recommendation domains (i.e., one-hot-encoded vectors of item preferences), these algorithms tend to have large input and output dimensionalities that dominate their overall size. This makes them difficult to train, due to the limited memory of graphical processing units, and difficult to deploy on mobile devices with limited hardware. To address these difficulties, we propose Bloom embeddings, a compression technique that can be applied to the input and output of neural network models dealing with sparse high-dimensional binary-coded instances. Bloom embeddings are computationally efficient, and do not seriously compromise the accuracy of the model up to 1/5 compression ratios. In some cases, they even improve over the original accuracy, with relative increases up to 12%. We evaluate Bloom embeddings on 7 data sets and compare it against 4 alternative methods, obtaining favorable results. We also discuss a number of further advantages of Bloom embeddings, such as 'on-the-fly' constant-time operation, zero or marginal space requirements, training time speedups, or the fact that they do not require any change to the core model architecture or training configuration.