Soft Computing: Introduction
What is computing?
Before understanding what is Soft computing we must know what is computing. In simple words, computing means mapping the given set of inputs to output using a formal method or an algorithm to solve a problem. In the context of computing, this input is called ‘Antecedent’ and the output is called ‘Consequent’.
The method of computing must be unambiguous, accurate, and provide a precise solution. Computing is suitable for problems that are easy to model mathematically.
Now, before moving on to soft computing let us understand what is hard computing and why there was a need to develop soft-computing algorithms.
Lotfi .A. Zade was the first person to introduce the concept of hard computing in 1996. According to him, a computing concept comes under the category of hard computing if :
>It provides precise results.
>Algorithm used to solve a problem is unambiguous.
>The control action is formally defined using an algorithm or a mathematical model.
Problems like finding derivatives, integrals, searching & sorting algorithms, finding the shortest path between two points, and many more for which we can get a precise result by using a mathematical model, comes under hard computing.
Need of Soft-Computing
Some of the real-world tasks like handwriting recognition, image classification, music generation, etc do not have an algorithm for computation of exact solutions in polynomial time. This is where Softcomputing comes into play.
The term Soft-computing was also coined by Lotfi. A. Zade [1]. He defined soft computing as follows:
Soft computing is a collection of methodologies that aim to exploit the tolerance of imprecision and uncertainity to achieve tractability, robustness, and low solution cost. Its principle components are fuzzy logic, neuro-computing and probabilistic reasoning. The role model for soft computing is human mind.
In the above definition, the following are a few key terms which one must understand:
- tolerance of imprecision: the result obtained using soft-computing is not precise.
- uncertainty: the soft-computing algorithm may give different results every time for the same problem.
- robustness: soft-computing algorithms can tackle any kind of input noise
- low solution cost: soft-computing makes it feasible to solve some of the problems which could be computationally very expensive if solved using hard computing.
Soft-computing algorithms are based on the biological decision-making system and follow methodologies like genetics, evolution, Ant’s behaviors, partical swarming, human nervous system etc. The three computing paradigms followed by soft-computing computations are fuzzy logics, neuro-computing, and probabilistic reasoning(genetic algorithm)[2].
Applications of Soft Computing
- Image processing
- Data Compression
- Fuzzy Logic Control
- Automative systems and Manufacturing
- Neuro-fuzzy systems
- Decision-support systems
and many more.
References
[1]Zadeh, L. A. (1996). Soft computing and fuzzy logic. In Fuzzy Sets, Fuzzy Logic, and Fuzzy Systems: Selected Papers by Lotfi a Zadeh (pp. 796–804).
[2]Li, X., Ruan, D., & Van der Wal, A. J. (1998). Discussion on soft computing at FLINS’96. International Journal of Intelligent Systems, 13(2‐3), 287–300.
