Learning Trans-Dimensional Random Fields with Applications to Language Modeling

30 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Learning Neural Trans-Dimensional Random Field Language Models with Noise-Contrastive Estimation (2017, methodology): “Different with the TRFs defined in [5, 6], a reference distribution q(x) is introduced as the baseline distribution, and the role of TRFs is to fit the difference between the data distribution and the reference distribution.”
Improved Training Of Neural Trans-Dimensional Random field Language Models with Dynamic Noise-Contrastive Estimation (2018, methodology): “TRF LMs are first proposed in [3] and further presented in [4] with thorough theoretical analysis and systematic evaluation.”
Residual Energy-Based Models for Text (2021, methodology): “Methods attempting at directly maximizing the likelihood of the data using gradient descent usually require sampling from the distribution induced by the model (Hinton, 2002b; Carreira-Perpinan and Hinton, 2005; Du and Mordatch, 2019; Xie et al., 2019; Wang et al., 2017).”
Language modeling with neural trans-dimensional random fields (2017, methodology): “The undirected graphical modeling approach, has few priori work, except [4, 5, 9].”