The fictions of Indian movies " Love Story 2050" and "Robot" are no longer dreams.These dreams have came true.Genetic programming is useful in all the fields like robotics, building construction , network security, mechanical engineering etc.
There is a hierarchy in machine learning systems: Genetic Algorithm(GA) is advancement of Artificial intelligence(AI) and Genetic Programming(GP) is advancement of GA.
GP is mainly based on Darwin's Theory of Evolution , Which tells about the Things change according to the environment. and Things evolve to be better.
GP simply creates gens like animals for machines and trains them and so the robots also behave like living things by making their decisions by their own.Simply the machine is given some data and ask it to learn from those data. GP is given problem, and then it formulate the problem and defines the best algorithm to solve that problem by itself.The programming language LISP is used in GP.
for example, In robotic soccer team, Each robot is given the data that it is supposed to follow the ball. But with using GP robot automatically learns the rules of soccer and on every different situations robot itself find the solution by which it can reach to the destination with ball.
To evolve the program following things are needed :
1. Population : Based on intial population the program can be evolved and it finds out which way is the best solution for given problem.
2. Fitness Function : It is the way to find out which program or which organ is best suited for given problem .
3. Breeding : beside the population and fitness function we need to improve the program . Breeding the program means extracting the best of the program using new generation technology
GP is now a days used for network security too. The old definitions of intrusion detection systems are fed as data by which GP predicts new definitions of attacks and finds the solution to stop such future attack. So that we can have the solution of the problem before actually it occurs, and we can stop zero day attack also.
GP has wide scope for researchers so researcher should go for learning genetic programming.
Good Luck!!
There is a hierarchy in machine learning systems: Genetic Algorithm(GA) is advancement of Artificial intelligence(AI) and Genetic Programming(GP) is advancement of GA.
GP is mainly based on Darwin's Theory of Evolution , Which tells about the Things change according to the environment. and Things evolve to be better.
GP simply creates gens like animals for machines and trains them and so the robots also behave like living things by making their decisions by their own.Simply the machine is given some data and ask it to learn from those data. GP is given problem, and then it formulate the problem and defines the best algorithm to solve that problem by itself.The programming language LISP is used in GP.
for example, In robotic soccer team, Each robot is given the data that it is supposed to follow the ball. But with using GP robot automatically learns the rules of soccer and on every different situations robot itself find the solution by which it can reach to the destination with ball.
To evolve the program following things are needed :
1. Population : Based on intial population the program can be evolved and it finds out which way is the best solution for given problem.
2. Fitness Function : It is the way to find out which program or which organ is best suited for given problem .
3. Breeding : beside the population and fitness function we need to improve the program . Breeding the program means extracting the best of the program using new generation technology
GP is now a days used for network security too. The old definitions of intrusion detection systems are fed as data by which GP predicts new definitions of attacks and finds the solution to stop such future attack. So that we can have the solution of the problem before actually it occurs, and we can stop zero day attack also.
GP has wide scope for researchers so researcher should go for learning genetic programming.
Good Luck!!
Comments
Post a Comment