Physics machine is a software which does physics like a physicist. First, all the things
human being has developed so far, for doing physics, will be available to this machine.
Secondly, all the consistent theories, successful experiments, and published articles will
be included into this machine in the form of traced and readable knowledge, in other
words, this machine can read and understand these all. Finally, as the last target, this
machine can observe the universe and physical events with the aim of creating theories
and physical laws.
METU, Physics Department, 06800, Ankara, Turkey
Keywords: Artificial Intelligence, Machine Learning, Data Mining, Artificial Physicist
There are approximately millions of articles over physics, huge collection of
very successful theories, and physics books. In the earth, no physicist could have
attempted to read and understand these accumulations since it requires free
infinite time, free space, infinite energy, and feeding from also other disciplinaries
and fields. Especially, philosophy, science history and mathematics are needed
in a very high understanding. Therefore, the first purpose of this thesis, which
is to inject all the needs to do physics, appears here to tackle with these vital
humanity problems. Hence, this research can be thought as an intention to meet
all these impossibilities.
Problems of Humanity
Human being has the age limit which is approximately 120; however intellectual
people are aging faster than others. Of course, this handicaps the productive
physicists from doing much more physics, especially in the very bright years of
her/him . Therefore, the need for this physics machine seems to one obvious
since it is immortal. On the other hand, this machine is not only older but
also more productive than humanity. It consists, recombines and drives artificial
physicists of all over the universe (no need for limited to the earth), physics
knowledge, and physical productions. Then, this machine can operate like an
artificial physical institution. Probably, this machine will be able to give us the
key to develop memory transfer applicable to humanity. Thus, the subsequent
mission of this physics machine will be the biology machine.
So far in the history of humanity, unfortunately, no one, no state, no institution
could have brought these things regarding physics together in the real world
and this seems to be impossible for the future humanity because there is no
attemption to solve this vital problem.
Knowledge is infinite. No one person or even very powerful states, e.g.
US, CCCP, or EU can own these infinite knowledge. For this, humanity needs
a huge machine. Therefore, this physics machine can be considered to be an
initiative huge machine.
Humanity is like a balloon which is filled with untraced and unreadable
knowledge gas. However, the developments especially in computer science and
philosophy has the ability of giving the key to structure and invent this software
machine. Some of the important fields in computer science which can be used
in the construction of this physics machine are Artificial Intelligence, Machine
Learning, and Data Mining. In the following lines, the researcher will analyze
these fields so that they can be used to implement this machine.
Artificial intelligence is a type of intelligence which has been the subject of not
only computer science but also philosophy. It can be regarded as the intelligence
of nonliving organisms. According to the researcher, since the organic intelligence
of human being, the institutions, the universities, or the states have no attempt
to mine and gather physics knowledge to establish much more vital physical
laws for superhumanity; the inorganic and nonliving intelligence, that is artificial
intelligence is the best candidate to operate this machine.
Like humanity and its efforts to learn so far in the education systems, this physics
machine is ready to learn and grow. Learning in computer is easier than that in
humanity since to trace and store the knowledge, transmit information, sorting
and memory transfer are faster and permanent than those in human brain. The
developments in machine learning field are used to do all these processes so
that this physics machine can operate the knowledge and practice to learn new
things in physics.
The most important thing to construct a theory in physics is to collect data as
the result of intense physical observations. In addition, to extract the correct
and usable data has vital importance. Data mining, then, will become the other
important operator to construct the physics machine.
To gather human physicists together in order to construct advanced machines or
theories is very difficult in this labor division system all over the world. Physics
and physicist are divided into million parts in terms of field and this is the
largest handicap to establish a large physical laws. This is because they cannot
combine their knowledge to produce new things. Even after this automation of
their knowledge and experiences, they leave with their previous knowledge and
practices, so they do not learn new things. Thus, artificial physicist can solve this
division problem in the construction of the physics machine.