Service-oriented applications commonly involve high-dimensional and sparse (HiDS) interactions among users and service-related entities, e.g., user-item interactions from a personalized recommendation services system. How to perform precise and efficient representation learning on such HiDS interactions data is a hot yet thorny issue. An efficient approach to it is latent factor analysis (LFA), which commonly depends on large-scale non-convex optimization. Hence, it is vital to implement an LFA model able to approximate second-order stationary points efficiently for enhancing its representation learning ability. However, existing ones suffer from high computational cost. To address this issue, this paper presents a Momentum-accelerated Hessian-vector algorithm (MH) for precise and efficient LFA on HiDS data. Its main ideas are two-fold: a) adopting the principle of a Hessian vector-product-based method to utilize the second-order information without manipulating a Hessian matrix directly, and b) incorporating a generalized momentum method into its parameter learning scheme for accelerating its convergence rate to a stationary point. Experimental results on nine industrial datasets demonstrate that compared with state-of-the-art LFA models, an MH-based LFA model achieves gains in both accuracy and convergence rate. These positive outcomes also indicate that a generalized momentum method is compatible with the algorithms, e.g., a second-order algorithms, e.g., a second-order algorithm, which implicitly rely on gradients.

Cloud computing is becoming an increasingly popular platform for the execution of scientific applications such as scientific workflows. In contrast to grids and other traditional high-performance computing systems, clouds provide a customizable infrastructure where scientific workflows can provision desired resources ahead of the execution and set up a required software environment on virtual machines (VMs). Nevertheless, various challenges, especially its quality-of-service prediction and optimal scheduling, are yet to be addressed. Existing studies mainly consider workflow tasks to be executed with VMs having time-invariant, stochastic, or bounded performance and focus on minimizing workflow execution time or execution cost while meeting the quality of service requirements. This work considers timevarying performance and aims at minimizing the execution cost of workflow deployed on Infrastructureas- a-Service clouds while meeting the constraints on Service-Level-Agreement. We employ time-seriesbased approaches to capture dynamic performance fluctuations, feed a genetic algorithm with predicted performance of VMs, and generate schedules at run-time. A case study based on real-world third-party IaaS clouds and some well-known scientific workflows shows that our proposed approach outperforms traditional approaches, especially those considering time-invariant or bounded performance only.

High-dimensional and incomplete (HDI) data are frequently encountered in big date-related applications for describing restricted observed interactions among large node sets. How to perform accurate and efficient representation learning on such HDI data is a hot yet thorny issue. A latent factor (LF) model has proven to be efficient in addressing it. However, the objective function of an LF model is nonconvex. Commonly adopted first-order methods cannot approach its second-order stationary point, thereby resulting in accuracy loss. On the other hand, traditional second-order methods are impractical for LF models since they suffer from high computational costs due to the required operations on the objective’s huge Hessian matrix. In order to address this issue, this study proposes a generalized Nesterov-accelerated second-order LF (GNSLF) model that integrates twofold conceptions: 1) acquiring proper second-order step efficiently by adopting a Hessian-vector algorithm and 2) embedding the second-order step into a generalized Nesterov’s acceleration (GNA) method for speeding up its linear search process. The analysis focuses on the local convergence for GNSLF’s nonconvex cost function instead of the global convergence has been taken; its local convergence properties have been provided with theoretical proofs. Experimental results on six HDI data cases demonstrate that GNSLF performs better than state-of-the-art LF models in accuracy for missing data estimation with high efficiency, i.e., a second-order model can be accelerated by incorporating GNA without accuracy loss.

Precise representation to undirected weighted network (UWN) is the foundation of understanding connection patterns inside a massive node set. It can be addressed via a Symmetric Non-negative Latent Factor (SNLF) model with a non-convex learning objective. However, existing SNLF models commonly adopt a first-order learning algorithm that cannot well handle such a non-convex objective, thereby leading to inaccurate UWN representation. Aiming at addressing this issue, this study incorporates an efficient second-order learning algorithm into an SNLF model, thereby establishing a Second-order Symmetric Non-negative Latent Factor (S ² NLF) model with two-fold ideas: a) applying the single latent factor-related mapping function to the non-negativity constrained optimization parameters to achieve an unconstrained learning objective, and b) optimizing this learning objective with its optimization parameters through an efficient second-order learning algorithm to achieve accurate representation to the target UWN with affordable computational burden. Empirical studies indicate that owing to its efficient incorporation of the second-order optimization technique, the proposed S ² NLF model outperforms state-of-the-art SNLF models when they are used to gain highly accurate representation to UWNs emerging from real applications.

Dear Editor, This letter presents a distributed adaptive second-order latent factor (DAS) model for addressing the issue of high-dimensional and incomplete data representation. Compared with first-order optimizers, a second-order optimizer has stronger ability in approaching a better solution when dealing with the non-convex optimization problems, thus obtaining better performance in extracting the latent factors (LFs) well representing the known information from high-dimensional and incomplete data. However, a traditional second- order optimizer are inefficient in exploiting the curvature information of an LF model due to its large number of parameters. In order to reduce the computational overhead, an inexact second-order method relying on the Hessian-free optimization is preferred. However, this method requires careful coordination of its components, which is time-consuming and impractical for real applications. To address the above issues, the DAS model leverages the curvature information with a Hessian-vector-incorporated inexact second-order optimizer and embeds it into a distributed, multi-phase, and multi-elitist learning particle swarm optimization (DM ² PSO) framework for efficient hyper-parameters adaptation and performance gain. Experimental results demonstrate that the DAS model outperforms several state-of- the-art models in estimating missing data on several high-dimensional and incomplete datasets from real-world applications.