The Bridge - Issue 2, 2018 - 6

Feature

Outsmart Moore's Law
with Machine Learning
by: Dustin Tanksley and Donald C. Wunsch

Over the last half century, computing has
transformed most aspects of society due to a rapid
increase in computation power. With the possible
end of Moore's Law in sight, much of this growth
could come to an end. This paper will discuss why
machine learning will continue growing even after
Moore's Law, and demonstrate why it is a great time
to enter the field.

WHAT IS MACHINE LEARNING?
At the most fundamental level, machine learning
is the process of using advanced function
approximators and large amounts of data to create
a mathematical representation of a problem. As
an example, one could take a group of pictures
of cats and dogs, and identify them. To do this,
some function would have to map these pictures
to a numeric value, possibly 0 for cats and 1 for
dogs. The art of machine learning is creating these
complicated functions and then apply meaning
to the mathematical representations. Three major
approaches to such tasks are supervised learning,
reinforcement learning, and unsupervised learning.
Supervised learning is typically the most intuitive.
In this type of learning, labeled data are fit to
an appropriate function; for example, matching
the price of a house to the size, other factors.
For simple problems, a linear (or higher order)
regression algorithm works just fine, however as
more parameters such as; bedroom and bathroom
count, location within or distance to a major city,
population density, and other factors are considered,
nonlinear models are often needed. Nonlinear
versions of regression exist, but neural networks and
other methods are often competitive with those,
and the whole family of such approaches can be

THE BRIDGE

considered types of machine learning. Perhaps the
best example of supervised learning success can be
seen in ImageNet, an effort to identify the object
in an image. ImageNet consists of more than 10
million internet images that have been identified
and labeled by humans, and as of 2017, the best
algorithm achieved a 97.7% correct classification,
which is better than most humans (typically 9095%) [1][2].
Reinforcement learning typically uses similar neural
network architectures as supervised learning,
with the key difference that data usually must be
generated/gathered, so that performance functions
replace the role of labels. Reinforcement learning
is easiest to visualize in games, such as tic-tac-toe.
Several actions are available, allowing the agent to
place a mark in one of the squares, and in doing
so generates a new data-point, but the agent does
not know if this was a good or bad move. When
the game ends, the agent is told if it wins, loses, or
draws and must use this data to determine if all the
actions it took were good or bad. Classifying moves
as good or bad is a somewhat difficult process, but
many improvements have been made in the field,
with the most recent being AlphaGo, an effort by
Google to master the game of Go (a far greater
computational challenge than Chess). AlphaGo's
most recent achievements include beating a former
world champion in a 5-game match (AlphaGo Lee)
and beating 60 of the top Go players in the world
without any losses (AlphaGo Master). Beyond
this, an even stronger version has been released,
AlphaGo Zero, which defeated AlphaGo Master
89-11 and is notable for being completely trained
by reinforcement learning from playing, without the
benefit of any initial supervised learning [3][4].
Unsupervised Learning is very different from
supervised and reinforcement learning, in that
it does not have a training target. Clustering, the
most common form of unsupervised learning,
groups inputs together based on similarities, and



Table of Contents for the Digital Edition of The Bridge - Issue 2, 2018

Contents
The Bridge - Issue 2, 2018 - Cover1
The Bridge - Issue 2, 2018 - Cover2
The Bridge - Issue 2, 2018 - Contents
The Bridge - Issue 2, 2018 - 4
The Bridge - Issue 2, 2018 - 5
The Bridge - Issue 2, 2018 - 6
The Bridge - Issue 2, 2018 - 7
The Bridge - Issue 2, 2018 - 8
The Bridge - Issue 2, 2018 - 9
The Bridge - Issue 2, 2018 - 10
The Bridge - Issue 2, 2018 - 11
The Bridge - Issue 2, 2018 - 12
The Bridge - Issue 2, 2018 - 13
The Bridge - Issue 2, 2018 - 14
The Bridge - Issue 2, 2018 - 15
The Bridge - Issue 2, 2018 - 16
The Bridge - Issue 2, 2018 - 17
The Bridge - Issue 2, 2018 - 18
The Bridge - Issue 2, 2018 - 19
The Bridge - Issue 2, 2018 - 20
The Bridge - Issue 2, 2018 - 21
The Bridge - Issue 2, 2018 - 22
The Bridge - Issue 2, 2018 - 23
The Bridge - Issue 2, 2018 - 24
The Bridge - Issue 2, 2018 - 25
The Bridge - Issue 2, 2018 - 26
The Bridge - Issue 2, 2018 - 27
The Bridge - Issue 2, 2018 - 28
The Bridge - Issue 2, 2018 - 29
The Bridge - Issue 2, 2018 - 30
The Bridge - Issue 2, 2018 - 31
The Bridge - Issue 2, 2018 - 32
The Bridge - Issue 2, 2018 - 33
The Bridge - Issue 2, 2018 - 34
The Bridge - Issue 2, 2018 - 35
The Bridge - Issue 2, 2018 - 36
The Bridge - Issue 2, 2018 - 37
The Bridge - Issue 2, 2018 - 38
The Bridge - Issue 2, 2018 - 39
The Bridge - Issue 2, 2018 - 40
The Bridge - Issue 2, 2018 - 41
The Bridge - Issue 2, 2018 - 42
The Bridge - Issue 2, 2018 - 43
The Bridge - Issue 2, 2018 - 44
https://www.nxtbook.com/nxtbooks/ieee/bridge_issue2_2022
https://www.nxtbook.com/nxtbooks/ieee/bridge_issue1_2022
https://www.nxtbook.com/nxtbooks/ieee/bridge_issue3_2021
https://www.nxtbook.com/nxtbooks/ieee/bridge_issue2_2021
https://www.nxtbook.com/nxtbooks/ieee/bridge_issue1_2021
https://www.nxtbook.com/nxtbooks/ieee/bridge_2020_issue3
https://www.nxtbook.com/nxtbooks/ieee/bridge_2020_issue2
https://www.nxtbook.com/nxtbooks/ieee/bridge_2020_issue1
https://www.nxtbook.com/nxtbooks/ieee/bridge_2019_issue3
https://www.nxtbook.com/nxtbooks/ieee/bridge_2019_issue2
https://www.nxtbook.com/nxtbooks/ieee/bridge_2019_issue1
https://www.nxtbook.com/nxtbooks/ieee/bridge_2018_issue3
https://www.nxtbook.com/nxtbooks/ieee/bridge_2018_issue2
https://www.nxtbook.com/nxtbooks/ieee/bridge_2018_issue1
https://www.nxtbookmedia.com