Adaptive Training of Random Mapping for Data Quantization

TL;DR

This paper introduces an adaptive training quantization method to improve data encoding quality by learning an optimal transform for random mapping, addressing limitations of cosine-based methods.

Contribution

It proposes a novel adaptive learning approach for binary embedding that enhances data pattern preservation in random quantization.

Findings

Outperforms existing random quantization methods in experiments.

Effectively preserves original data information in reduced codes.

Addresses the limitations of cosine-based random mapping.

Abstract

Data quantization learns encoding results of data with certain requirements, and provides a broad perspective of many real-world applications to data handling. Nevertheless, the results of encoder is usually limited to multivariate inputs with the random mapping, and side information of binary codes are hardly to mostly depict the original data patterns as possible. In the literature, cosine based random quantization has attracted much attentions due to its intrinsic bounded results. Nevertheless, it usually suffers from the uncertain outputs, and information of original data fails to be fully preserved in the reduced codes. In this work, a novel binary embedding method, termed adaptive training quantization (ATQ), is proposed to learn the ideal transform of random encoder, where the limitation of cosine random mapping is tackled. As an adaptive learning idea, the reduced mapping is adaptively calculated with idea of data group, while the bias of random transform is to be improved to hold most matching information. Experimental results show that the proposed method is able to obtain outstanding performance compared with other random quantization methods.