Adaptive Regularization for Similarity Measures

نویسندگان

  • Koby Crammer
  • Gal Chechik
چکیده

Algorithms for learning distributions over weight-vectors, such as AROW (Crammer et al., 2009) were recently shown empirically to achieve state-of-the-art performance at various problems, with strong theoretical guaranties. Extending these algorithms to matrix models pose challenges since the number of free parameters in the covariance of the distribution scales as n with the dimension n of the matrix, and n tends to be large in real applications. We describe, analyze and experiment with two new algorithms for learning distribution of matrix models. Our first algorithm maintains a diagonal covariance over the parameters and can handle large covariance matrices. The second algorithm factors the covariance to capture inter-features correlation while keeping the number of parameters linear in the size of the original matrix. We analyze both algorithms in the mistake bound model and show a superior precision performance of our approach over other algorithms in two tasks: retrieving similar images, and ranking similar documents. The factored algorithm is shown to attain faster convergence rate.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Optimized co-registration method of Spinal cord MR Neuroimaging data analysis and application for generating multi-parameter maps

Introduction: The purpose of multimodal and co-registration In MR Neuroimaging is to fuse two or more sets images (T1, T2, fMRI, DTI, pMRI, …) for combining the different information into a composite correlated data set in order to visualization, re-alignment and generating transform to functional Matrix. Multimodal registration and motion correction in spinal cord MR Neuroimag...

متن کامل

Regularization with Adaptive Neighborhood Condition for Image Denoising

Image denoising by minimizing a similarity of neighborhoodbased cost function is presented. This cost function consists of two parts, one related to data fidelity and the other is a structure preserving smoothing term. The latter is controlled by a weight coefficient that measures the neighborhood similarity between two pixels and attaching an additional term penalizes it. Unlike most work in n...

متن کامل

Point Similarity Measure Based on Mutual Information

Registration of multi-modality images requires similarity measures that can deal with complex and unknown image intensity dependencies. Such measures have to rely on statistics, and consequently, they require relatively large image regions to operate. This makes the detection of localized image discrepancies difficult. As a solution we propose point similarity measures, which can measure simila...

متن کامل

SOME SIMILARITY MEASURES FOR PICTURE FUZZY SETS AND THEIR APPLICATIONS

In this work, we shall present some novel process to measure the similarity between picture fuzzy sets. Firstly, we adopt the concept of intuitionistic fuzzy sets, interval-valued intuitionistic fuzzy sets and picture fuzzy sets. Secondly, we develop some similarity measures between picture fuzzy sets, such as, cosine similarity measure, weighted cosine similarity measure, set-theoretic similar...

متن کامل

Simultaneously Learning Neighborship and Projection Matrix for Supervised Dimensionality Reduction

Explicitly or implicitly, most of dimensionality reduction methods need to determine which samples are neighbors and the similarity between the neighbors in the original highdimensional space. The projection matrix is then learned on the assumption that the neighborhood information (e.g., the similarity) is known and fixed prior to learning. However, it is difficult to precisely measure the int...

متن کامل

An Empirical Comparison of Distance Measures for Multivariate Time Series Clustering

Multivariate time series (MTS) data are ubiquitous in science and daily life, and how to measure their similarity is a core part of MTS analyzing process. Many of the research efforts in this context have focused on proposing novel similarity measures for the underlying data. However, with the countless techniques to estimate similarity between MTS, this field suffers from a lack of comparative...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2012