Reconfigurable Inverted Index
Reconfigurable Inverted Index (Rii) is a research project and open-source library for approximate nearest neighbor and similarity search over high-dimensional vectors. It focuses on flexible, reconfigurable inverted index structures that support efficient vector search, making it directly relevant as a vector-search engine component for AI and multimedia retrieval applications.
About this tool
Reconfigurable Inverted Index (Rii)
Category: SDKs & Libraries
Tags: ANN, vector-indexing, similarity-search
Website: https://yusukematsui.me/project/rii/
Overview
Reconfigurable Inverted Index (Rii) is an open-source research library for approximate nearest neighbor (ANN) and similarity search over high-dimensional vectors. It is built on top of the IVFADC framework and is designed to efficiently support searches over both entire databases and arbitrary subsets, while maintaining performance as new items are added.
Features
-
Approximate Nearest Neighbor Search
- Supports fast similarity search over high-dimensional vectors.
- Performance comparable to state-of-the-art systems such as Faiss (per the referenced experiments).
-
Reconfigurable Inverted Index Structure
- Based on the standard IVFADC system.
- Uses a specially designed data layout where items are stored linearly.
-
Efficient Subset Search
- Enables efficient search over arbitrary subsets of the database.
- Can switch from inverted-index search to a linear product-quantization (PQ) scan when the target subset is small, improving efficiency for constrained queries.
-
Dynamic Updates and Reconfiguration
- Data structure can be dynamically adjusted after adding many new items.
- Maintains fast query speed even after large-scale insertions, mitigating typical performance degradation seen in many ANN systems.
-
Research-Backed Design
- Algorithm and data structure are described and evaluated in a peer-reviewed ACMMM 2018 paper.
- Includes extensive experimental comparisons with existing ANN systems.
Academic Reference
If you use Rii in academic work, you can cite:
@inproceedings{acmmm_matsui_2018,
author={Yusuke Matsui and Ryota Hinami and Shin'ichi Satoh},
title={Reconfigurable Inverted Index},
booktitle={ACM International Conference on Multimedia (ACMMM)},
pages={1715--1723},
year={2018}
}
Pricing
Rii is presented as a research project and open-source library; no pricing or commercial plans are specified on the provided content.
Loading more......
Information
Categories
Tags
Similar Products
6 result(s)AiSAQ is an all-in-storage approximate nearest neighbor search system that uses product quantization to enable DRAM-free vector similarity search, serving as a specialized vector search/indexing approach for large-scale information retrieval.
Ruby gem for approximate nearest neighbor search that can integrate with pgvector and other backends to power vector similarity search in Ruby applications.
EFANNA is an extremely fast approximate nearest neighbor search algorithm based on kNN graphs and randomized KD-trees. The provided implementation offers a high-performance ANN index suitable as a building block in custom vector search and retrieval infrastructure.
iRangeGraph is an ANN indexing approach and accompanying implementation for range-filtering nearest neighbor search. It provides a specialized graph-based index that supports vector similarity search under range constraints, making it directly useful as a component or reference implementation for advanced vector database indexing and retrieval.
NSG is an approximate nearest neighbor search algorithm based on a sparse navigable graph structure designed for high-dimensional vector similarity search. The reference implementation provides a graph-based ANN index that can be integrated into custom vector retrieval systems.
Product-Quantization is a GitHub repository implementing the inverted multi-index structure for product-quantization-based approximate nearest neighbor search, providing building blocks for scalable vector search engines.