aiqbench analysis
This research introduces entropy-punctured Bloom Filters as a novel compression technique for machine learning features, offering substantial storage savings while maintaining predictive fidelity in regression tasks. By removing low-variability bit positions, the method outperforms classical approaches like PCA and Random Projection under memory-constrained conditions without requiring raw data access. The findings provide a practical solution for optimizing model efficiency in bandwidth-limited or privacy-sensitive environments.
Entropy-Punctured Bloom Filters for Memory-Efficient Machine Learning
arXiv:2609.14187v1 Announce Type: cross Abstract: Memory-efficient feature representations are increasingly important in machine learning settings where storage, transmission cost, bandwidth, or privacy constraints limit access to raw data. Bloom Filter (BF) encodings provide compact probabilistic representations of engineered features, but their behavior under structural compression and their applicability to regression tasks remain underexplored. In this work, we propose entropy-punctured Bloom Filters, a memory-aware encoding strategy that removes low-variability bit positions identified using empirical entropy. Starting from fixed-length BF encodings of quantized features, the proposed approach produces reduced representations that preserve predictive structure while improving predictive efficiency relative to encoded representation size. We evaluate the approach on diverse regression datasets, comparing raw features, Principal Component Analysis (PCA), Random Projection (RP), and Bloom Filter variants under leakage-free evaluation protocols and approximately matched representation sizes. Performance is assessed using ridge regression, XGBoost, and neural networks, with predictive efficiency measured as R2 relative to encoded representation size per sample. Results show that Bloom Filter encodings remain competitive with classical compressed representations while achieving substantial storage savings. Entropy-based puncturing further reduces representation size with minimal loss in predictive fidelity, yielding improved predictive efficiency. These findings demonstrate that entropy-punctured Bloom Filters provide an effective representation-level compression approach for memory-constrained machine learning.
Abstract:Memory-efficient feature representations are increasingly important in machine learning settings where storage, transmission cost, bandwidth, or privacy constraints limit access to raw data. Bloom Filter (BF) encodings provide compact probabilistic representations of engineered features, but their behavior under structural compression and their applicability to regression tasks remain underexplored. In this work, we propose entropy-punctured Bloom Filters, a memory-aware encoding strategy that removes low-variability bit positions identified using empirical entropy. Starting from fixed-length BF encodings of quantized features, the proposed approach produces reduced representations that preserve predictive structure while improving predictive efficiency relative to encoded representation size. We evaluate the approach on diverse regression datasets, comparing raw features, Principal Component Analysis (PCA), Random Projection (RP), and Bloom Filter variants under leakage-free evaluation protocols and approximately matched representation sizes. Performance is assessed using ridge regression, XGBoost, and neural networks, with predictive efficiency measured as R2 relative to encoded representation size per sample. Results show that Bloom Filter encodings remain competitive with classical compressed representations while achieving substantial storage savings. Entropy-based puncturing further reduces representation size with minimal loss in predictive fidelity, yielding improved predictive efficiency. These findings demonstrate that entropy-punctured Bloom Filters provide an effective representation-level compression approach for memory-constrained machine learning.
Accepted for publication in the Proceedings of the 38th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2026). 9 pages, 7 figures
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2609.14187 [cs.LG]
(or arXiv:2609.14187v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2609.14187 arXiv-issued DOI via DataCite (pending registration)
Full text under the publisher's open licence · 244 words
Related coverage
Cascaded Learned Bloom Filter for Optimizing Model-Filter Size Balance and Fast Rejection
arXiv · Aug 25, 2026
News Archive 2026
Quantum Computing Report · Jul 25, 2026
PySpark Optimization: 12 Proven Techniques to Speed Up Your Spark Jobs
Analytics Vidhya · May 27, 2026
News
Quantum Computing Report · Jun 7, 2026