Questions 1. What is co-operative machine learning in multi-agent environment? In such case how two different types of a

[answerhappygod]

Questions 1. What is co-operative machine learning in multi-agent environment? In such case how two different types of agents can learn together selectively? Design multi-agent system with co-operative learning for medicine delivery in a hospital. In this case prescribed medicines to be delivered to a particular patient room within half an hour. What will be function of different agents in this case? What will be PEAS for these agents? How ‘Best first search’ algorithm can be used in this case. Can we use Euclidean distance in this case to determine heuristic values?
2. You have time stamped performance indices of different machines. Can you use CNN to detect future performance and optimal maintenance schedule? How? Model it as a constraint satisfaction problem that handles time series data with time series forecasting methods.
3. A football coach decides to keep track of players’ movements during different time slots in the match and cluster them on field playing style of them to form batches for training. Further they want to enhance the algorithm for the team selection. The plan is to use Kmeans clustering, CNN and model-based agents for the same. Design the agent – Discuss PEAS. Discuss details of working of agent function. Also discuss what data would you sense for the same.
4. What is a learning agent? How it learns in association with the environment? Take an example where you want to develop an intelligent system to decide the best way for user to spend next one hour in the city considering his/her location. Social media data of the user along with information of his/her current location along with entertainment avenues are used for input data. Can you use Table based agents in this case? What would be limitations of this type of solution? How we can embed more intelligence using goalbased and utility-based agents?
5. Design a fitness function using conditional probability for selection of friends on social media. These different friends compete for friendship and then there is multi-point cross over and mutation to create new expected behavior of the friend. Design a detailed genetic algorithm based approach for the same.
6. What is activation function in neural networks? What are different types of neural networks? What are different types of activation functions? Explain detailed working of CNN with one application. How can you improve filtering in this case if you are looking for developing CNN to recommend a match to watch to a team before their actual encounter with certain team?
7. Discuss heuristic and non-heuristic search methods? Discuss A* search and iterative deepening A* search algorithms with their advantages and disadvantages. How it can be used for selection in genetic algorithm? What could be fitness function in this case? Discuss this approach for product recommendation.
8. Can my route selection to office be modelled as a constraint satisfaction problem? How? What is constraint satisfaction problem (CSP)? Give three examples of CSP. Write a detail algorithm with example. Discuss algorithms to deal with such problems with one detailed worked out example.
9. What is local beam search? What is greedy algorithm? What is hill climbing search? What are different variants of hill climbing search? Can we use random restart greedy algorithm for selection in genetic algorithm? How? Use this approach for developing an agent helping to select an insurance policy.