TRIBHUVAN UNIVERSITY
INSTITUTE OF SCIENCE AND TECHNOLOGY
2082 B.S.
Candidates are required to give their answers in their own words as far as practicable.
Group A
Long Answer QuestionsAttempt any Two questions.
- 1What are the problems of depth limited search? Use the Alpha-Beta pruning algorithm to prune the following game tree assuming child nodes are visited from left to right. Show all final alpha and beta values computed at root, each internal node explored, and at the top of pruned branches.
- 2How do you infer knowledge from semantic net? Illustrate with an example. Consider the following Bayesian network with random variables A, B, C and D. If A is true, find the probability of D being true.
- 3What is model free reinforcement learning? Differentiate between active and passive reinforcement learning. Describe the different operators used in genetic algorithm.
Group B
Short Answer QuestionsAttempt any Eight questions.
- 4How do you define rational? Can AI choose between right and wrong? Justify.
- 5What do you mean by PEAS description? Give an example.
- 6Apply the hill climbing search algorithm to solve the following problem. Assume the heuristic function h(n)= +1 for all blocks in the support structure if the block is correctly positioned, otherwise -1
- 7Why do we need posterior probability? Design a semantic network for the following statements. 1. Dogs hate cats. 2. Tom is a cat. 3. Puppy is a dog. 4. Cats chase rats. 5. Rats are clever.
- 8Design a Hebb net to implement logical OR function.
- 9Given the premises: all living things are either animal or plant. Plant needs sunlight for photosynthesis. Mustard is a living thing but it is not an animal. Does mustard need sunlight for photosynthesis?
- 10What are the phases for developing an expert system? Explain.
- 11Why do machines need to understand and generate natural language? What might be the challenges? Explain.
- 12List any one example of static and dynamic environment for an agent. Why and when do you prefer rule based systems?