Machine learning to predict numerical outcomes in a matrix-defined problem space

A non-transitory computer-readable medium storing instructions which, when implemented by processing circuitry of one or more computers, cause the processing circuitry to perform operations comprising:

• accessing a matrix, the matrix having multiple dimensions, one dimension of the matrix representing features, and one dimension of the matrix representing data points;
• separating the matrix into multiple submatrices along a first dimension, each submatrix including all cells in the matrix for a set of values in the first dimension;
• providing the multiple submatrices to multiple machines;
• computing, using each machine, a correlation between values in at least one second dimension of the matrix and a value for a preselected feature in the matrix, the correlation being used to predict the value for the preselected feature based on other values along the at least one second dimension;
• and
• providing an output representing the computed correlation.

A method comprising:

• accessing a matrix, the matrix having multiple dimensions, one dimension of the matrix representing features, and one dimension of the matrix representing data points;
• separating the matrix into multiple submatrices along a first dimension, each submatrix including all cells in the matrix for a set of values in the first dimension;
• providing the multiple submatrices to multiple machines;
• receiving, from each machine, a correlation between values in at least one second dimension of the matrix and a value for a preselected feature in the matrix;
• and
• generating an output representing the computed correlation.