Stochastic Non-convex Ordinal Embedding with Stabilized Barzilai-Borwein Step Size

TL;DR

This paper introduces SVRG-SBB, a scalable stochastic algorithm for non-convex ordinal embedding that avoids SVD computations and adaptively chooses step sizes, demonstrating efficiency and effectiveness in large-scale data.

Contribution

The paper proposes a novel SVD-free stochastic algorithm with stabilized Barzilai-Borwein step size for non-convex ordinal embedding, improving scalability and convergence.

Findings

Achieves convergence to a stationary point at rate O(1/T).

Demonstrates lower computational cost compared to state-of-the-art methods.

Shows good prediction performance on real-world data.

Abstract

Learning representation from relative similarity comparisons, often called ordinal embedding, gains rising attention in recent years. Most of the existing methods are batch methods designed mainly based on the convex optimization, say, the projected gradient descent method. However, they are generally time-consuming due to that the singular value decomposition (SVD) is commonly adopted during the update, especially when the data size is very large. To overcome this challenge, we propose a stochastic algorithm called SVRG-SBB, which has the following features: (a) SVD-free via dropping convexity, with good scalability by the use of stochastic algorithm, i.e., stochastic variance reduced gradient (SVRG), and (b) adaptive step size choice via introducing a new stabilized Barzilai-Borwein (SBB) method as the original version for convex problems might fail for the considered stochastic \textit{non-convex} optimization problem. Moreover, we show that the proposed algorithm converges to a stationary point at a rate $\mathcal{O}(\frac{1}{T})$ in our setting, where $T$ is the number of total iterations. Numerous simulations and real-world data experiments are conducted to show the effectiveness of the proposed algorithm via comparing with the state-of-the-art methods, particularly, much lower computational cost with good prediction performance.