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.

This paper addresses the problem of efficiently estimating posterior tree-structured graphical models (e.g., Gaussian Processes, Kernel Models and Kernel Bayesian Networks). The main challenge is to obtain a sufficiently large posterior of unknown state, which is a crucial metric for many graphical models. In this work, we proposed a stochastic optimization problem, and present an efficient algorithm that is the equivalent of minimizing the sum of the sum of the regularized and the nonconvex regular functions. We first consider the problem of stochastic estimation, and show that it is NP-hard: we give a stochastic optimization algorithm that is significantly more tractable in terms of solving a series of random steps in a finite time. To this end, we present an efficient approximation of the algorithm to the linear model of our paper, with the goal of overcoming a number of the drawbacks. In particular, we show that using an estimator-based estimator as the baseline for the stochastic estimation algorithm is not feasible. We thus propose an adaptive stochastic optimization algorithm for estimation.

On-device Scalable Adversarial Reasoning with MIMO Feedback

An investigation into the use of color channel filters in digital image watermarking

# The Evolution of Lexical Variation: Does Language Matter?

Unsupervised Domain Adaptation for Object Detection

An Online Convex Relaxation for Kernel Likelihood EstimationThis paper addresses the problem of efficiently estimating posterior tree-structured graphical models (e.g., Gaussian Processes, Kernel Models and Kernel Bayesian Networks). The main challenge is to obtain a sufficiently large posterior of unknown state, which is a crucial metric for many graphical models. In this work, we proposed a stochastic optimization problem, and present an efficient algorithm that is the equivalent of minimizing the sum of the sum of the regularized and the nonconvex regular functions. We first consider the problem of stochastic estimation, and show that it is NP-hard: we give a stochastic optimization algorithm that is significantly more tractable in terms of solving a series of random steps in a finite time. To this end, we present an efficient approximation of the algorithm to the linear model of our paper, with the goal of overcoming a number of the drawbacks. In particular, we show that using an estimator-based estimator as the baseline for the stochastic estimation algorithm is not feasible. We thus propose an adaptive stochastic optimization algorithm for estimation.