A Momentum-Incorporated Non-Negative Latent Factorization of Tensors Model for Dynamic Network Representation
This is an incremental improvement for dynamic network representation in big data applications.
The paper tackled the problem of poor tail convergence in stochastic gradient descent-based latent factorization of tensors models for dynamic networks by proposing a momentum-incorporated non-negative model, which achieved higher prediction accuracy and convergence speed on two datasets.
A large-scale dynamic network (LDN) is a source of data in many big data-related applications due to their large number of entities and large-scale dynamic interactions. They can be modeled as a high-dimensional incomplete (HDI) tensor that contains a wealth of knowledge about time patterns. A Latent factorization of tensors (LFT) model efficiently extracts this time pattern, which can be established using stochastic gradient descent (SGD) solvers. However, LFT models based on SGD are often limited by training schemes and have poor tail convergence. To solve this problem, this paper proposes a novel nonlinear LFT model (MNNL) based on momentum-incorporated SGD, which extracts non-negative latent factors from HDI tensors to make training unconstrained and compatible with general training schemes, while improving convergence accuracy and speed. Empirical studies on two LDN datasets show that compared to existing models, the MNNL model has higher prediction accuracy and convergence speed.