HCLAE: High Capacity Locally Aggregating Encodings for Approximate Nearest Neighbor Search

TL;DR

This paper introduces HCLAE, a novel encoding method for approximate nearest neighbor search that improves quantization accuracy and search efficiency, with an online learning scheme and a new search structure called A-Tree.

Contribution

The paper proposes HCLAE and Dictionary Annealing for better data encoding, along with A-Tree for efficient search, advancing the state-of-the-art in ANN methods.

Findings

Lower quantization error than existing methods

Significant speed-up in ANN search tasks

Effective online learning scheme for large-scale data

Abstract

Vector quantization-based approaches are successful to solve Approximate Nearest Neighbor (ANN) problems which are critical to many applications. The idea is to generate effective encodings to allow fast distance approximation. We propose quantization-based methods should partition the data space finely and exhibit locality of the dataset to allow efficient non-exhaustive search. In this paper, we introduce the concept of High Capacity Locality Aggregating Encodings (HCLAE) to this end, and propose Dictionary Annealing (DA) to learn HCLAE by a simulated annealing procedure. The quantization error is lower than other state-of-the-art. The algorithms of DA can be easily extended to an online learning scheme, allowing effective handle of large scale data. Further, we propose Aggregating-Tree (A-Tree), a non-exhaustive search method using HCLAE to perform efficient ANN-Search. A-Tree achieves magnitudes of speed-up on ANN-Search tasks, compared to the state-of-the-art.