Natural computation is inspired by nature. All branches in natural computation are intended to find the solution for the problems within nature. The three branches are computation inspired by nature, the simulation and emulation of nature, and computation with natural materials. All this branches allow the development of novel computation tools whether in software, hardware or 'wetware' for problems solving. Hence, natural computation is very important field of research.
In the middle of 20th centuries, many researchers interested in Natural Computation, most of them try to find out what is natural computation; how it work, how it done, and it is applicable or not. But now, natural computation is the most importance field in these 21st centuries. Firstly, what it is means by Natural computation? The word natural itself is define as existing in or formed in nature; not made or caused by human intervention. While the word computation in general is defines as any type of process, algorithm or measurement; it is not limited to digital data only but also includes phenomena from the way human think until calculation with more specific meanings. Computation also can be defined as a process of a well-defined model that can be understood and can be expressed in protocol, algorithm, network topology and more.
Natural computation or also called as natural computing, is field of research that works with computational techniques that depend on or inspired from nature and natural systems, which authorize the development of new computational tools for problems solving, which guide to the fusion natural patterns, behaviors, and organisms, and may result in the design of new computational systems that use natural medium to compute (de Castro LN. 2006). The computational approaches of natural computation are based on the most simplified version of the mechanisms and processes that present in the related natural phenomena.
The research in natural computation is genuinely interdisciplinary field and it is related with experimental issues, applied issues and theoretical issues. The fact is natural computation form a connection between information, sciences and nature.
The natural computation can be divided into three main branches, which are computing that inspired by nature for the development of new problems solving systems, the simulation and emulation of nature by means of computing, and computing with natural materials.
The first branch is computing that inspired by nature for development of new problems solving systems. This branch of Natural computing applied the nature as motivation for the development of problems solving systems. The main idea of this branch is to develop algorithm tools so that we can compute the input by taking the idea from nature for the solution of complex problems. By that, we can simplify the complex problems into simple language for human understanding.
Next branch is the simulation and emulation of nature by means of computing. It is basically an artificial process which focuses at modeling patterns, forms, behaviors, and organisms. Its output can be used to mimic numerous types of natural phenomena, resulting in improving our understanding of nature and close about computer models.
Last branch is computing with natural materials. It relate to the use of new natural materials to execute computation, hence signify a true new computing paradigm that comes to replace or complement the current silicon-based computers.
Therefore, natural computation can be defined as the research field that based on or inspired by nature that allows the development of new computational tools; software, hardware or 'wetware', for problems solving, which guide to the fusion of natural patterns, behaviors, and organisms resulting in the designing new computing systems that use nature as medium to compute.
Natural computation is very important field these days. Most of applications, systems, and devices are implemented with natural computation concept. Natural computation is important to human beings because peoples need something that can make their life easier. By using natural computing, all human work load can be reduced, lesser time needed to done something, and reduce workforce needed.
The importances of natural computation are differing between branches. Each branches have own functions, aim, and goals.
3.1 Computing inspired by nature
Among all natural computations branches, computational algorithm and systems that inspired from nature are the oldest and the most popular topics. They come up with two main ideas, modeling natural phenomena and simulate in computer, and involving the study of natural phenomena, procedure and theoretical models for the advancement of computational systems and algorithms with ability to solve complex problems.
The intentions of these ideas are to invent theoretical models, which can be applied in computer, and to supply optional solution techniques to the problems that could not be solved by other traditional techniques, such as linear, non-linear, and dynamic programming. Among many approaches in this branch; computation inspired by nature, the most popular among researcher are the artificial neural network, evolutionary algorithms, swarm intelligence and artificial immune systems.
3.2 The simulation and emulation of nature by means of computation
The second branch of natural computation, come out with tools for the simulation and emulation of nature, and study of natural phenomena that can be used to test biological theories that cannot be tested with traditional experiment and analysis techniques. There are two main ideas in the simulation and emulation on nature in computers; use tools for studying the fractal geometry of nature or use artificial life techniques.
3.3 Computation with natural materials
This third and last branch of Natural computation is concerned with new methods of computational algorithm based on other natural materials other than silicon. The computer technology changes very fast, from gears to relays to valve to transistors to integrated circuits. In this modernize world, now we can have a single silicon chips with millions of logic gates. This efficiency of the most basis information processing elements is unavoidably going to achieve a state where logic gates will be very small like an atom. Computing with natural materials, cause a major change in the current computing technology.
Enthused by the need of finding alternative medium for computing, researchers are now attempting to design new computers based on molecules, such as Deoxyribonucleic acid (DNA) and Ribonucleic acid (RNA) or quantum theory. These ideas now resulted what is known as molecular computing such as DNA computing, and quantum computation such as quantum mechanics.
Hence, natural computation is very important field. The advancement in this natural computation field will bring advantages to human. This field also can make the human-nature relation become more understandable and easier. By natural computation we learn about nature and how can we implemented into understandable language.
4.0 When Natural Computation is Use?
There are many types of problems that need to be solve, natural phenomena to be synthesized and question about nature to be answered. in most cases, one or more natural computation branches will be used in problems solving, synthesize the phenomenon or answer the questions. Yet, natural computation branches not only act as tools for problems solving, but also not always the suitable and efficient branches. In other situation where alternate solution techniques is used for a problems could be provide superior results.
Natural computation should be used when there are complex problems that alternative solution could not solve such as the problems involves a large number of variables, highly dynamic, nonlinear, has multiple objectives, and more. Next is the problems to be solved could not be modeled, such as pattern recognition and classification task. However, in some cases, even though it could not be modeled, there are samples available that can be used to guide the system to solve the problems.
Other situation when natural computation should be used is when a single solution cannot solve or it is not good enough or when diversity is important. Most standard problems-solving techniques are able to give a single solution but not able to give the solution more than one. Besides that, it is used when biological, physical, and chemical systems and processes have to be synthesized with realism.
Natural computation also used when life behaviors and phenomena have to be synthesized in artificial medium. The core of a provided natural behavior or pattern is extracted and synthesized usually in a much simpler form in artificial life systems.
Last but not least, natural computation is used when the limitations of current technology are reached or new computing materials have to be sought. Nature flourish with information and processing on how to use these natural materials to compute are the main challenges of the third branch of natural computation, computation with natural materials.
Natural computation has many area of study such as Cultural algorithms, DNA computation, Neural computation, Quantum computation, Artificial life and more. The idea of Natural computation is to mimic the complex phenomena that happening in the nature as a computational processes for improving the computation method that is executed mainly from a problems solving. All this study is still undergo transformation by transformation that important by trying to fit into computational studies that using the ideas from the nature all around.
5.1 Cultural algorithms
Cultural algorithms are computational model in cultural growth. It is dividing into two elements, a population space; by using evolutionary algorithms, and a belief space. These two elements related by means of a vote-inherit-promote protocol. The information obtain from the problems solving activity of the population can be stored in the belief space in the form of production rules. Cultural algorithms representing general structure for generate hybrid evolutionary systems that combine evolutionary search and domain knowledge.
5.2 DNA computation
Deoxyribo-Nucleic Acid (DNA) is used in this DNA computing instead of the traditional silicon based microprocessors. The encoding of DNA is just the same with string of binary data which it is encoded with one and zero, while DNA encoded with four bases which represented with A, T, C, and G. The DNA property that important is it has double stranded nature with every DNA sequence having a natural complement. This complementarily features because DNA become unique data structure for computation and can be used in many ways (Chen K. 2004).
5.3 Neural Computation
In this study, Artificial neural networks are one of the Neural Computation. Artificial neural network have been developed as generalizations of mathematical models of biological nervous systems. The outcome of the synapses are stand by weight that modulate the output of associate input signals and the non linear feature show by neurons is represented by a transfer function, which is usually the sigmoid, Gaussian function and more. The neuron stimulus then calculated as the weighted sum of the input signals, transformed by the function transfer. Artificial neuron can learn by adjusting the weights in agreement to the chosen learning algorithm. The neural network learning situation could be classified into three different types, that are, supervised learning, unsupervised learning and reinforcement learning.
5.4 Quantum computation
The measurement in quantum computer is different with conventional silicon computer that is in conventional silicon computer, the amount of data is measured by bits while in a quantum computer, and it is measured in qubits. Qubits is quantum bits. A qubit can be a 1 or a 0 or it can be presented in a superposition that is simultaneously both 1 and 0 or somewhere in between. The fundamental of quantum computation is that the properties of quantum particles can be used in representing and structuring data and devised quantum mechanisms can be used in performing operation with this data (Nielsen MA, Chuang IL, 2000).
5.5 Artificial life
Artificial life or also called as alife is an efforts in setting up systems with life which has properties like biological organisms had, such as homeostasis, adaptability, reproduction and more. Alife is always expressed as effort in understanding high level behavior from low level rules. By understanding this bond in specific systems resulted to give new solution to complex real-world problems, such as stock market prediction, data mining on the internet and disease prevention.
There is still having other areas of study. All of the areas are basically nature based study. Human need to understand the nature so that human can know what can be done and what cannot be done.
Natural computation has trained peoples to think naturally about computation and also think computationally about nature. Natural computation approaches produce new and exciting capabilities for science as a total, mostly computer science, engineering, biosciences and philosophy. The products of these discoveries are constantly becoming novel technological solutions and explanations to old and latest problems.
Since last three decades, natural computing has undergone an astounding development. Its boundary has been developing in many directions simultaneously, from nature-inspired computing to the emerging computing with new natural materials. This fast development is cause by enlargement of computing power and memory storage capacity have increased dramatically in this decades and the enormous quantity of problems with features such as nonlinearity, high-dimensionality, difficulty in modeling and finding derivatives, impracticable traditional techniques such as linear, nonlinear, dynamic, integer programming and others, and growth in relation and cooperation between different disciplines and fields of research such as computer science, engineering, physics, ecology, biology, medicine, mathematics, neuroscience and statistic.
The applications of natural computation are diverse, and in last few years, the growth of importance application in optimizing problems in general, biological application, data mining applications, games, biometrics, robotics, economics, forecasting problems and application involving the security of information systems. The natural computation gives the problems solving system in human life. Therefore, natural computation needs to be learning so that human can understand nature.