Data Compression and Statistical Inference

Statistical Inference Through Data Compression

Text Compression and Language Modeling

Data compression is bound up with grammatical inference and prediction. If we can infer a grammar for some data, we may use that grammar to compress that data and to predict what data is likely to occur in the future. In this paper I describe how the popular text compression techniques actually work. I then compare the results achieved by the ad-hoc dictionary methods with the modern statistica...

Data Compression and Its Statistical Implications, with an Application to the Analysis of Microarray Images

Data Compression and Its Statistical Implications, with an Application to the Analysis of Microarray Images This thesis consists of three parts. Even though each part is self-contained, a common theme runs through all of them: data compression and its implications for statistical inference. In particular, we consider the following three questions. How can we quantify the eeect of compression on...

Compression Trade-Offs for Multiple Inference Tasks in Sensor Networks

Introduction Advances in sensor and communication technology have focused interest on using networks of sensors. Sensor node tasks include collecting data, sharing the data, collectively making inferences (estimations and decisions) from the data, and communicating the collected data and/or the inference results to one or more information sinks where further inferences may be made. Sensor netwo...

Inference by Conversion

We are discussing a modeling technique based on the idea to generate data sequences with a number of suggested models. These sequences are transformed, or converted, into an observed data sequence by a suitable function, or a program. The motivation for doing so is in cases where the likelihood of observed data is hard to compute, which is circumvented with an indirect approximation by trying t...