Fast Latent Variable Models for Inference and Visualization on Mobile Devices

TL;DR

This paper introduces Vedalia, a distributed network system that enables fast inference and visualization of latent variable models on mobile devices, specifically enhancing Amazon review analysis with improved performance and reduced server costs.

Contribution

The paper presents RLDA, a new latent variable model extending LDA with auxiliary data, and a distributed system that offloads computation to mobile devices for efficient inference.

Findings

Rapid inference on mobile devices with minimal server resources

Enhanced modeling accuracy using auxiliary review data

Significant reduction in server costs and response times

Abstract

In this project we outline Vedalia, a high performance distributed network for performing inference on latent variable models in the context of Amazon review visualization. We introduce a new model, RLDA, which extends Latent Dirichlet Allocation (LDA) [Blei et al., 2003] for the review space by incorporating auxiliary data available in online reviews to improve modeling while simultaneously remaining compatible with pre-existing fast sampling techniques such as [Yao et al., 2009; Li et al., 2014a] to achieve high performance. The network is designed such that computation is efficiently offloaded to the client devices using the Chital system [Robinson & Li, 2015], improving response times and reducing server costs. The resulting system is able to rapidly compute a large number of specialized latent variable models while requiring minimal server resources.