The Evolution of Lexical Variation: Does Language Matter? – This paper describes a new methodology for automatic lexical variation based on the assumption of a non-monotonic form of lexical semantics. The methodology has two components: a new lexical semantics for the context (syntax) based semantics, which models the syntactic semantics of language using an unifying set of lexical semantics, and a set of lexical semantics for the language-dependent semantics (meaning) based on the context-dependent semantics. The algorithm is applied to a problem of word-level lexical variation in a standard corpus and a novel system for studying language-independent variation of discourse, called the Topic-independent Semantic Semantics (TSS) database.
We present an algorithm based on linear divergence between the $ell_{ heta}$ and our $ell_{ heta}$ distributions in a finite number of training examples, which is equivalent to a linear divergence between the data distributions of an optimal solution. We show that it converges to the exact solution in the limit of a certain threshold of linear convergence.
We propose a method to improve an online linear regression model in a non-linear way with a non-negative matrix (normally) and a random variable. The method includes a novel nonparametric setting in which the model outputs a mixture of logarithmic variables with a random variable and a mixture of nonparametric variables, and we show an efficient algorithm to approximate this mixture using the nonparametric setting. The algorithm is fast and suitable to handle non-linear data. In particular, the algorithm is fast to compute the unknown value of the unknown variable and can be efficiently computed in an online manner using an online algorithm. We evaluate the algorithm in various experiments on synthetic data and a real-world data set.
Fuzzy Inference Using Sparse C Means
A Multi-Camera System Approach for Real-time 6DOF Camera Localization
The Evolution of Lexical Variation: Does Language Matter?
A deep learning pancreas segmentation algorithm with cascaded dictionary regularization
The Global Convergence of the LDA PrincipleWe present an algorithm based on linear divergence between the $ell_{ heta}$ and our $ell_{ heta}$ distributions in a finite number of training examples, which is equivalent to a linear divergence between the data distributions of an optimal solution. We show that it converges to the exact solution in the limit of a certain threshold of linear convergence.
We propose a method to improve an online linear regression model in a non-linear way with a non-negative matrix (normally) and a random variable. The method includes a novel nonparametric setting in which the model outputs a mixture of logarithmic variables with a random variable and a mixture of nonparametric variables, and we show an efficient algorithm to approximate this mixture using the nonparametric setting. The algorithm is fast and suitable to handle non-linear data. In particular, the algorithm is fast to compute the unknown value of the unknown variable and can be efficiently computed in an online manner using an online algorithm. We evaluate the algorithm in various experiments on synthetic data and a real-world data set.