Method and apparatus for learning to classify patterns and assess the value of decisions

Abstract

An apparatus and method for training a neural network model (21) to classify patterns (26) or to assess the value of decisions associated with patterns by comprising the actual output of the network in response to an input pattern with the desired output for that pattern on the basis of a Risk Differential Learning (RDL) objective function (28), the results of the comparison governing adjustment of the neural network model's parameters by numerical optimization. The RDL objective function includes one or more terms, each being a risk / benefit / classification figure-of-merit (RBCFM) function, which is a synthetic, monotonically non-decreasing, anti-symmetric / asymmetric, piecewise-differentiable function of a risk differential (Fig. 6), which is the difference between outputs of the neural network model produced in response to a given input pattern. Each RBCFM function has mathematical attributes such that RDL can make universal guarantees of maximum correctness / profitability and minimum complexity. A strategy for profit-maximizing resource allocation utilizing RDL is also disclosed.

Description

technical field [0001] The present application relates to statistical pattern recognition and / or classification, and more particularly to learning strategies whereby a strategy computer can learn how to identify and recognize concepts. Background technique [0002] Pattern recognition and / or classification are useful in a wide variety of real-world tasks, such as those related to optical feature recognition, interpretation of remote sensing imaging, medical diagnosis / decision support, digital telecommunication, and the like. Such pattern recognition is typically achieved by trainable networks such as neural networks, which through a series of training exercises are able to "learn" the concepts necessary to perform the pattern classification task. Such networks are trained by feeding them (a) learned instances with concepts of interest represented mathematically by an ordered set of numbers, referred to here as "input patterns", and (b) respectively associated with these Ins...

AI Technical Summary

Problems solved by technology

[0006] In practice, however, it has been found that in many practical examples differential learning as described above fails to provide the aforementioned guarantees

Likewise, the differential learning concept places specific needs on the learning process in relation to the nature of the data being learned, as well as limitations on the mathematical characteristics of the adopted neural network representative model that affect pattern classification

Moreover, the previous analysis of differential learning only dealt with pattern classification, leaving no mention of another type of problem about value estimation, that is, profit and loss prediction based on input pattern estimation decisions (enumerated by the output of neural network patterns).

Images