Kernel Measures of Independence for non-iid Data
نویسندگان
چکیده
Many machine learning algorithms can be formulated in the framework of statistical independence such as the Hilbert Schmidt Independence Criterion. In this paper, we extend this criterion to deal with structured and interdependent observations. This is achieved by modeling the structures using undirected graphical models and comparing the Hilbert space embeddings of distributions. We apply this new criterion to independent component analysis and sequence clustering.
منابع مشابه
A Kernel Statistical Test of Independence
Although kernel measures of independence have been widely applied in machine learning (notably in kernel ICA), there is as yet no method to determine whether they have detected statistically significant dependence. We provide a novel test of the independence hypothesis for one particular kernel independence measure, the Hilbert-Schmidt independence criterion (HSIC). The resulting test costs O(m...
متن کاملA Recursive Approximation Approach of non-iid Lognormal Random Variables Summation in Cellular Systems
Co-channel interference is a major factor in limiting the capacity and link quality in cellular communications. As the co-channel interference is modeled by lognormal distribution, sum of the co-channel interferences of neighboring cells is represented by the sum of lognormal Random Variables (RVs) which has no closed-form expression. Assuming independent, identically distributed (iid) RVs, the...
متن کاملGaussian Process Subset Scanning for Anomalous Pattern Detection in Non-iid Data
Identifying anomalous patterns in real-world data is essential for understanding where, when, and how systems deviate from their expected dynamics. Yet methods that separately consider the anomalousness of each individual data point have low detection power for subtle, emerging irregularities. Additionally, recent detection techniques based on subset scanning make strong independence assumption...
متن کاملKernel bandwidth estimation for non-parametric density estimation: a comparative study
We investigate the performance of conventional bandwidth estimators for non-parametric kernel density estimation on a number of representative pattern-recognition tasks, to gain a better understanding of the behaviour of these estimators in high-dimensional spaces. We show that there are several regularities in the relative performance of conventional kernel bandwidth estimators across differen...
متن کاملتشخیص سرطان پستان با استفاده از برآورد ناپارمتری چگالی احتمال مبتنی بر روشهای هستهای
Introduction: Breast cancer is the most common cancer in women. An accurate and reliable system for early diagnosis of benign or malignant tumors seems necessary. We can design new methods using the results of FNA and data mining and machine learning techniques for early diagnosis of breast cancer which able to detection of breast cancer with high accuracy. Materials and Methods: In this study,...
متن کامل