Dynamic across-time measurement interpretation

Proceedings of the Eighth National Conference on Artificial Intelligence (AAAI-90), Boston, MA, July 1990
Dynamic across-time measurement interpretation
Dennis DeCoste

Incrementally maintaining a qualitative understanding of physical system behavior based on observations is crucial to real-time process monitoring, diagnosis, and control.

This paper describes the DATMI theory for dynamically maintaining a pinterp-space, a concise representation of the local and global interpretations consistent with observations over time. Each interpretation signifies an alternative path of states in a qualitative envisionment.

DATMI can use domain-specific knowledge about state and transition probabilities to maintain the best working interpretation. By maintaining the space of alternative interpretations as well, DATMI avoids the need for extensive backtracking to handle incomplete or faulty data.

Another publication from the same category: Machine Learning and Data Science

IEEE Computing Conference 2018, London, UK

Regularization of the Kernel Matrix via Covariance Matrix Shrinkage Estimation

The kernel trick concept, formulated as an inner product in a feature space, facilitates powerful extensions to many well-known algorithms. While the kernel matrix involves inner products in the feature space, the sample covariance matrix of the data requires outer products. Therefore, their spectral properties are tightly connected. This allows us to examine the kernel matrix through the sample covariance matrix in the feature space and vice versa. The use of kernels often involves a large number of features, compared to the number of observations. In this scenario, the sample covariance matrix is not well-conditioned nor is it necessarily invertible, mandating a solution to the problem of estimating high-dimensional covariance matrices under small sample size conditions. We tackle this problem through the use of a shrinkage estimator that offers a compromise between the sample covariance matrix and a well-conditioned matrix (also known as the "target") with the aim of minimizing the mean-squared error (MSE). We propose a distribution-free kernel matrix regularization approach that is tuned directly from the kernel matrix, avoiding the need to address the feature space explicitly. Numerical simulations demonstrate that the proposed regularization is effective in classification tasks.