Abstract:
Statistical Learning is the process of estimating an unknown probabilistic inputoutput
relationship of a system using a limited number of observations; and a statistical
learning machine (SLM) is the machine that learned such a process. While their roots
grow deeply in Probability Theory, SLMs are ubiquitous in the modern world.
Automatic Target Recognition (ATR) in military applications, Computer Aided
Diagnosis (CAD) in medical imaging, DNA microarrays in Genomics, Optical
Character Recognition (OCR), Speech Recognition (SR), spam email filtering, stock
market prediction, etc., are few examples and applications for SLM; diverse fields but
one theory.
The field of Statistical Learning can be decomposed to two basic subfields, Design
and Assessment. We mean by Design, choosing the appropriate method that learns from
the data to construct an SLM that achieves a good performance. We mean by
Assessment, attributing some performance measures to the designed SLM to assess this
SLM objectively. To achieve these two objectives the field encompasses different other
fields: Probability, Statistics and Matrix Theory; Optimization, Algorithms, and
programming, among others.
Three main groups of specializations—namely statisticians, engineers, and computer
scientists (ordered ascendingly by programming capabilities and descendingly by
mathematical rigor)—exist on the venue of this field and each takes its elephant bite.
Exaggerated rigorous analysis of statisticians sometimes deprives them from
considering new ML techniques and methods that, yet, have no “complete"
mathematical theory. On the other hand, immoderate add-hoc simulations of computer
scientists sometimes derive them towards unjustified and immature results. A prudent
approach is needed that has the enough flexibility to utilize simulations and trials and
errors without sacrificing any rigor. If this prudent attitude is necessary for this field it is
necessary, as well, in other fields of Engineering.
In the spirit of this prelude, this article is intended to be a pilot-view of the field that
sheds the light on SLM applications, the Design and Assessment stages, necessary
mathematical and analytical tools, and some state-of-the-art references and research.