# Unsupervised Multi-Index Semantic Hashing

arXiv: https://arxiv.org/abs/2103.14460 · HTML: https://mlsanity.com/paper/2103.14460/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 35 references that are in the map.
- First posted on arXiv 26 Mar 2021.

## Builds on (most cited first, 14 of 14)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Auto-Encoding Variational Bayes](https://mlsanity.com/paper/1312.6114/) (2013)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [A Survey on Learning to Hash](https://mlsanity.com/paper/1606.00185/) (2016)
- [Learning to Hash for Indexing Big Data - A Survey](https://mlsanity.com/paper/1509.05472/) (2015)
- [Candidate Generation with Binary Codes for Large-Scale Top-N Recommendation](https://mlsanity.com/paper/1909.05475/) (2019)
- [Content-aware Neural Hashing for Cold-start Recommendation](https://mlsanity.com/paper/2006.00617/) (2020)
- [Variational Deep Semantic Hashing for Text Documents](https://mlsanity.com/paper/1708.03436/) (2017)
- [NASH: Toward End-to-End Neural Architecture for Generative Semantic Hashing](https://mlsanity.com/paper/1805.05361/) (2018)
- [Fast Exact Search in Hamming Space with Multi-Index Hashing](https://mlsanity.com/paper/1307.2982/) (2013)
- [Recurrent Binary Embedding for GPU-Enabled Exhaustive Retrieval from Billion-Scale Semantic Vectors](https://mlsanity.com/paper/1802.06466/) (2018)
- [Unsupervised Semantic Hashing with Pairwise Reconstruction](https://mlsanity.com/paper/2007.00380/) (2020)
- [Unsupervised Neural Generative Semantic Hashing](https://mlsanity.com/paper/1906.00671/) (2019)
- [Document Hashing with Mixture-Prior Generative Models](https://mlsanity.com/paper/1908.11078/) (2019)

## Built on by (most cited first, 5 of 5)

- [Hyperbolic Hierarchical Contrastive Hashing](https://mlsanity.com/paper/2212.08904/) (2022)
- [Representation Learning for Efficient and Effective Similarity Search and Recommendation](https://mlsanity.com/paper/2109.01815/) (2021)
- [Sequential Modelling with Applications to Music Recommendation, Fact-Checking, and Speed Reading](https://mlsanity.com/paper/2109.06736/) (2021)
- [Learning Similarity Preserving Binary Codes for Recommender Systems](https://mlsanity.com/paper/2204.08569/) (2022)
- [Bit-mask Robust Contrastive Knowledge Distillation for Unsupervised Semantic Hashing](https://mlsanity.com/paper/2403.06071/) (2024)

## Abstract (arXiv)

Semantic hashing represents documents as compact binary vectors (hash codes) and allows both efficient and effective similarity search in large-scale information retrieval. The state of the art has primarily focused on learning hash codes that improve similarity search effectiveness, while assuming a brute-force linear scan strategy for searching over all the hash codes, even though much faster alternatives exist. One such alternative is multi-index hashing, an approach that constructs a smaller candidate set to search over, which depending on the distribution of the hash codes can lead to sub-linear search time. In this work, we propose Multi-Index Semantic Hashing (MISH), an unsupervised hashing model that learns hash codes that are both effective and highly efficient by being optimized for multi-index hashing. We derive novel training objectives, which enable to learn hash codes that reduce the candidate sets produced by multi-index hashing, while being end-to-end trainable. In fact, our proposed training objectives are model agnostic, i.e., not tied to how the hash codes are generated specifically in MISH, and are straight-forward to include in existing and future semantic hashing models. We experimentally compare MISH to state-of-the-art semantic hashing baselines in the task of document similarity search. We find that even though multi-index hashing also improves the efficiency of the baselines compared to a linear scan, they are still upwards of 33% slower than MISH, while MISH is still able to obtain state-of-the-art effectiveness.