anytime algorithms for learning anytime classifiers saher ... - Technion

Technion - Computer Science Department - Ph.D. Thesis PHD-2008-12 - 2008
Chapter 1
Introduction
Machine learning techniques are gaining prevalence in the production of a wide
range of classifiers for complex real-world applications with nonuniform testing
and misclassification costs. The increasing complexity of these applications poses
a real challenge to resource management during learning and classification.
Assume, for example, that a medical center has decided to use machine learning
techniques to build a diagnostic tool for heart disease. Figure 1.1 shows three
possible trees. The first tree (upper-left) makes diagnoses using only the results of
cardiac catheterization (heart cath). This tree is expected to be highly accurate.
Nevertheless, the high costs and risks associated with the heart cath procedure
make this decision tree impractical. The second tree (lower-left) dispenses with
the need for cardiac catheterization and makes a diagnosis based on a single, simple,
inexpensive test: whether or not the patient complains of chest pain. Such
a tree would be highly accurate: most people do not experience chest pain and
are indeed healthy. The tree, however, does not distinguish between the costs of
different types of errors. While a false positive prediction might result in extra
treatments, a false negative prediction might put a person’s life at risk. Therefore,
a third tree (right) is preferred, one that attempts to minimize test costs
and misclassification costs simultaneously.
As another example, consider a hardware manufacturer who uses a machinelearning
based tool for assuring the quality of produced chips. In realtime, each
chip in the pipeline is scanned and several features can be extracted from the
image. The features vary in their computation time. The manufacturer trains
the component using thousands of chips whose validity is known. Because the
training is done offline, the manufacturer can provide the values of all possible
features, regardless of their computation time. In realtime, however, the model
must make a decision within 2 seconds. Therefore, for each chip, the classifier
may use features whose total computation time is at most 2 seconds. Unlike
the medical center, the manufacturer is not interested in minimizing the total
5