Survey On The Estimation Of Mutual Information Methods as a Measure of Dependency Versus Correlation Analysis

In this survey, we present and compare different approaches to estimate Mutual Information (MI) from data to analyse general dependencies between variables of interest in a system. We demonstrate the performance difference of MI versus correlation analysis, which is only optimal in case of linear dependencies. First, we use a piece-wise constant Bayesian methodology using a general Dirichlet prior. In this estimation method, we use a two-stage approach where we approximate the probability distribution first and then calculate the marginal and joint entropies. Here, we demonstrate the performance of this Bayesian approach versus the others for computing the dependency between different variables. We also compare these with linear correlation analysis. Finally, we apply MI and correlation analysis to the identification of the bias in the determination of the aerosol optical depth (AOD) by the satellite based Moderate Resolution Imaging Spectroradiometer (MODIS) and the ground based AErosol RObotic NETwork (AERONET). Here, we observe that the AOD measurements by these two instruments might be different for the same location. The reason of this bias is explored by quantifying the dependencies between the bias and 15 other variables including cloud cover, surface reflectivity and others.

Towards Identification of Relevant Variables in the observed Aerosol Optical Depth Bias between MODIS and AERONET observations

Measurements made by satellite remote sensing, Moderate Resolution Imaging Spectroradiometer (MODIS), and globally distributed Aerosol Robotic Network (AERONET) are compared. Comparison of the two datasets measurements for aerosol optical depth values show that there are biases between the two data products. In this paper, we present a general framework towards identifying relevant set of variables responsible for the observed bias. We present a general framework to identify the possible factors influencing the bias, which might be associated with the measurement conditions such as the solar and sensor zenith angles, the solar and sensor azimuth, scattering angles, and surface reflectivity at the various measured wavelengths, etc. Specifically, we performed analysis for remote sensing Aqua-Land data set, and used machine learning technique, neural network in this case, to perform multivariate regression between the ground-truth and the training data sets. Finally, we used mutual information between the observed and the predicted values as the measure of similarity to identify the most relevant set of variables. The search is brute force method as we have to consider all possible combinations. The computations involves a huge number crunching exercise, and we implemented it by writing a job-parallel program.

Maximum Joint Entropy and Information-Based Collaboration of Automated Learning Machines

We are working to develop automated intelligent agents, which can act and react as learning machines with minimal human intervention. To accomplish this, an intelligent agent is viewed as a question-asking machine, which is designed by coupling the processes of inference and inquiry to form a model-based learning unit. In order to select maximally-informative queries, the intelligent agent needs to be able to compute the relevance of a question. This is accomplished by employing the inquiry calculus, which is dual to the probability calculus, and extends information theory by explicitly requiring context. Here, we consider the interaction between two question-asking intelligent agents, and note that there is a potential information redundancy with respect to the two questions that the agents may choose to pose. We show that the information redundancy is minimized by maximizing the joint entropy of the questions, which simultaneously maximizes the relevance of each question while minimizing the mutual information between them. Maximum joint entropy is therefore an important principle of information-based collaboration, which enables intelligent agents to efficiently learn together.