---
title: Batch, match, and patch: low-rank approximations for score-based variational inference
url: https://www.ml-quant.com/papers/arxiv/2410.22292/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-09-26
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: arXiv:2410.22292
source_url: https://arxiv.org/abs/2410.22292
featured: 2024-10-31
citations: 9
topic: Other
---


# Batch, match, and patch: low-rank approximations for score-based variational inference

The paper introduces an extension of the batch-and-match framework for black-box variational inference to high-dimensional problems, using compact parameterization of full covariance matrices to enhance efficiency and performance.

- Source: https://arxiv.org/abs/2410.22292
- Identifier: arXiv:2410.22292
- Released: 2024-10-29
- First featured: Quant Letter No. 72 (2024-10-31): https://www.ml-quant.com/issues/2024-10-31/
- Citations (Semantic Scholar): 9
- Published in: International Conference on Artificial Intelligence and Statistics
- Topic: Other

## Related

- [State Equations in High-Dimensional Regression](https://www.ml-quant.com/papers/repec/taf-lstaxx-v-54-y-2025-i-3-p-850-863/): The article explores the interchangeability of various state equations in high-dimensional statistics through parameter transformations.
- [Interior-Point Linear SVMs](https://www.ml-quant.com/papers/repec/spr-joptap-v-202-y-2024-i-1-d-10-1007-s10957-022-02103-1/): The paper uses multiple variable splitting to solve binary classification and novelty detection problems in high-dimensional data, demonstrating competitive results against other methods and specific algorithms.
- [HighDimensional Data Privacy](https://www.ml-quant.com/papers/ssrn/5278724/): The article explores a new data analysis and protection method that merges principal component analysis with differential privacy for improved handling of high-dimensional data and privacy protection.
- [Hybrid Random Projection for Enhanced High-Dimensional Data Representation](https://www.ml-quant.com/papers/ssrn/4705621/): A novel hybrid random projection method, HBP, merges the benefits of normal random projection and plus-minus one random projection, enhancing data dimension reduction while maintaining data structure.
- [Depth Anything V2](https://www.ml-quant.com/papers/arxiv/2406.09414/): Depth Anything V2 is a new model for monocular depth estimation, using synthetic and large-scale pseudo-labeled real images for faster, more accurate results and setting a new evaluation benchmark.
- [MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark](https://www.ml-quant.com/papers/arxiv/2406.01574/): MMLU-Pro, an improved dataset, expands the Massive Multitask Language Understanding benchmark by adding tougher questions and more choices, serving as a better benchmark to monitor progress in the field.