Fast rates for empirical vector quantization
نویسندگان
چکیده
منابع مشابه
Fast rates for empirical vector quantization
Abstract: We consider the rate of convergence of the expected loss of empirically optimal vector quantizers. Earlier results show that the mean-squared expected distortion for any fixed distribution supported on a bounded set and satisfying some regularity conditions decreases at the rate O(logn/n). We prove that this rate is actually O(1/n). Although these conditions are hard to check, we show...
متن کاملAlgorithms for Fast Vector Quantization∗
Nearest neighbor searching is an important geometric subproblem in vector quantization. Existing studies have shown that the difficulty of solving this problem efficiently grows rapidly with dimension. Indeed, existing approaches on unstructured codebooks in dimension 16 are little better than brute-force search. We show that if one is willing to relax the requirement of finding the true neares...
متن کاملFast Encoding Algorithm for Vector Quantization
In this paper, we present a new and fast encoding algorithm (FEA) for vector quantization. The magnitude (sum of the components of a vector) feature of the vectors is used in this algorithm to improve the efficiency of searching. Sorting of the magnitude values enhances the searching. As the values are sorted, the searching can be terminated in advance to reduce the time needed to locate the re...
متن کاملFast Resampling Using Vector Quantization
We present a fast resampling scheme using vector quantiza-tion. Our method diiers from prior work applying vector quantization to speeding up image and volume processing in two essential aspects. First, our method uses blocks with overlapping rather than disjoint extents. Second, we present a means of trading oo smaller block sizes for additional computation. These two innovations allow vector ...
متن کاملVery Fast Tree-structured Vector Quantization
Very fast tree-structured vector quantization employs scalar quantization decisions at each level, but chooses the dimension on which to quantize based on the coordinate direction of maximum variance. Because the quantization is scalar, searches are no more complex than scalar quantization − providing significant improvement in complexity over full-searched or even tree-structured vector quanti...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Electronic Journal of Statistics
سال: 2013
ISSN: 1935-7524
DOI: 10.1214/13-ejs822