(Aktu Btech) Machine Learning Techniques Important Unit-5 Reinforcement Learning and Genetic Algorithm

B.Tech AKTU Quantum Book will take you on a journey through the world of Machine Learning Techniques. Access critical notes, frequently asked questions, and important insights to grasp this innovative profession. Unit-5 Reinforcement Learning and Genetic Algorithm

Dudes 🤔.. You want more useful details regarding this subject. Please keep in mind this as well.

Important Questions For Machine Learning Techniques:
*Quantum               *B.tech-Syllabus
*Circulars                *B.tech AKTU RESULT
* Btech 3rd Year     * Aktu Solved Question Paper

Q1. Describe reinforcement learning. 

Ans.

• 1. The study of reinforcement learning focuses on how creatures and artificial systems may adapt their behaviour in response to rewards and penalties.
• 2. Algorithms for reinforcement learning connected to dynamic programming, a generic method for optimal control.
• 3. Neurobiological studies of neuromodulation and addiction as well as psychology studies of animal behaviour have both shown evidence of reinforcement learning processes.
• 4. The goal of reinforcement learning is to learn an environment-specific optimal policy using observed rewards. A policy that maximizes the predicted total reward is said to be optimum.
• 5. The agent won’t have any basis for choosing what to do without some feedback on what is good and what is bad.
• 6. The agents must understand that when they win, something good has occurred, and when they lose, something negative has occurred.
• 7. This form of feedback is referred to as reinforcement or reward.
• 8. In the realm of robotics, where there are many tasks that are too complicated to be encoded as algorithms and no training data is available, reinforcement learning is useful.
• 9. Reinforcement learning is the only practical method for teaching a programme to perform at a high level in many difficult domains.

Q2. Explain different machine learning task.

Ans. Following are most common machine learning tasks:

• 1. Data preprocessing: Data preparation is crucial before beginning the model training process. As part of data preprocessing following is done: 
  • a. Data cleaning 
  • b. Handling missing data  
• 2. Exploratory data analysis: Exploratory data analysis is the next stage after preprocessing data in order to comprehend data distribution and relationships between/within the data.
• 3. Feature engineering / selection: One of the crucial processes involved in creating machine learning models is feature selection. The proper features must be chosen in order to produce models that are more accurate and to accomplish goals such as making models simpler to build and reducing overtitling, among others. 
• 4. Regression: The estimation of numerical values is the focus of regression activities (continuous variables). Examples include estimating the cost of homes, the cost of goods, the value of stocks, etc.
• 5. Classification: Choosing a category for a data set before classifying it is a classification task (discrete variables). The most typical examples include determining whether or not an email is spam, whether or not a person has a specific ailment, whether or not a transaction is fraudulent, etc.
• 6. Clustering: Finding natural groups of data and a name for each of these groups is the main goal of clustering activities (clusters). Customer segmentation and the identification of product features for product roadmaps are a few examples that are frequently used. 
• 7. Multivariate querying: Multivariate querying is about querying or finding similar objects.
• 8. Density estimation: Problems with density estimation are related to determining an object’s likelihood or frequency. 
• 9. Dimension reduction: Dimension reduction, also known as feature selection and feature extraction, is the process of minimizing the number of random variables taken into account.
• 10. Model algorithm/selection: There are frequently many models that have been trained using various algorithms. Selecting the best models to use in production is one of the crucial tasks.
• 11. Testing and matching: Testing and matching tasks relates to comparing data sets. 

Q3. Write short note on Genetic algorithm.

Ans.

• 1. Based on the principles of natural genetics and natural selection, genetic algorithms are computerized search and optimisation algorithms.
• 2. These algorithms develop search and optimisation procedures by imitating the natural selection and genetics principles.
• 3. The design space is transformed into the genetic space by genetic algorithms. A group of workable solutions is called the design space.
• 4. Genetic algorithms work with a coding of variables. 
• 5. The benefit of using a coding of variables space is that, even though the function may be continuous, coding discretizes the search space. 
• 6. Search space is the space for all possible feasible solutions of particular problem. 
• 7. Following are the benefits of Genetic algorithm: 
  • a. They are robust. 
  • b. They provide optimization over large space state. 
  • c. They do not break on slight change in input or presence of noise. 
• 8. Following are the application of Genetic algorithm: 
  • a. Recurrent neural network 
  • b. Mutation testing 
  • c. Code breaking 
  • d. Filtering and signal processing 
  • e. Learning fuzzy rule base  

Q4. Explain different phases of genetic algorithm.  

Ans. Different phases of genetic algorithm are: 

1. Initial population: 

• a. The process begins with a set of individuals which is called a population. 
• b. Each individual is a solution to the problem we want to solve. 
• c. An individual is characterized by a set of parameters (variables) known as genes. 
• d. Genes are joined into a string to form a chromosome (solution). 
• e. In a genetic algorithm, the set of genes of an individual is represented using a string. 
• f. Usually, binary values are used (string of 1s and 0s). 

2. FA (Factor Analysis) fitness function: 

• a. The fitness function gauges a person’s level of fitness (the ability of all individual to compete with other individual).
• b. It gives a fitness score to each individual. 
• c. The probability that an individual will be selected for reproduction is based on its fitness score. 

3. Selection: 

• a. The purpose of the selection phase is to choose the most fit people and allow them to pass on their genes to the following generation. 
• b. Two pairs of individuals (parents) are selected based on their fitness Scores. 
• c. Individuals with high fitness have more chance to be selected for reproduction.  

4. Crossover: 

• a. Crossover is the most significant phase in a genetic algorithm. 
• b. For each pair of parents to be mated, a crossover point is chosen at random from within the genes.
• c. For example, consider the crossover point to be 3 as shown:  

• d. Offspring are created by exchanging the genes of parents among themselves until the crossover point is reached. 

• e. The new offspring are added to the population. 

5. Mutation: 

• a. A modest random likelihood of mutation can occur in some of the genes of newly created children. 
• b. This implies that some of the bits in the bit string can be flipped. 

• c. Mutation occurs to maintain diversity within the population and prevent premature convergence. 

6. Termination: 

• a. If the population has converged, the algorithm is finished (does not produce offspring which are significantly different from the previous generation). 
• b. Then it is said that the genetic algorithm has provided a set of solutions to our problem.

Q5. Draw genetics cycle for genetic algorithm.

Ans. Generational cycle of GA:

Components of generational cycle in GA: 

• 1. Population (Chromosomes): A population is a grouping of people. A population is made up of the individuals being tested, their phenotypic parameters, and some knowledge of the search space.
• 2. Evaluation (Fitness): A fitness function is a specific kind of objective function that measures how optimal a solution in this case, a chromosome is in a genetic algorithm so that it can be compared to every other chromosome.
• 3. Selection: A portion of the current population is chosen to breed a new generation during each succeeding generation. The selection of individual solutions is dependent on fitness.
• 4. Generic operator: An operator called a genetic operator is one that a genetic algorithm uses to direct the algorithm towards a solution to a particular problem. 

Q6. What is the main function of crossover operation in genetic algorithm ?

Ans.

• 1. In a genetic algorithm, crossover is the fundamental operator. The crossover operator affects the genetic algorithm’s performance.
• 2. The crossover operator utilized for a given issue relies on the encoding scheme employed.
• 3. Extending the crossover points results in the genetic material from two parents being exchanged.

Function of crossover operation/operator in genetic algorithm: 

• 1. The primary purpose of a crossover operator is to increase population variety.
• 2. The performance of the genetic algorithm can be enhanced by specialized crossover created for a particular issue.
• 3. Crossover combines parental approaches to create offspring in the hopes of creating superior approaches.
• 4. In order to ensure proper mixing of the construction blocks, crossover operators are essential.
• 5. Exploration and exploitation are kept in balance via crossover. The effectiveness of genetic algorithms is a result of exploitation and exploration strategies. Exploitation is the process of using knowledge that already exists to identify a better solution, whereas exploration entails looking into novel and undiscovered solutions in an exploratory environment.

Machine Learning Techniques Btech Quantum PDF, Syllabus, Important Questions

Machine Learning Techniques Quantum PDF | AKTU Quantum PDF:

AKTU Important Links | Btech Syllabus

Important Links-Btech (AKTU)