Q#1: Knowledge in learning refers to:
(A) Information used by a learning algorithm to generalize
(B) Raw input data only
(C) Output predictions only
(D) Random guesses
Answer: (A) Information used by a learning algorithm to generalize
Q#2: Inductive bias is:
(A) The set of assumptions a learning algorithm uses to predict unseen examples
(B) Random noise in data
(C) BFS only
(D) DFS only
Answer: (A) The set of assumptions a learning algorithm uses to predict unseen examples
Q#3: Bias in learning helps to:
(A) Reduce the hypothesis space
(B) Increase noise
(C) BFS only
(D) DFS only
Answer: (A) Reduce the hypothesis space
Q#4: Too strong bias can lead to:
(A) Underfitting
(B) Overfitting
(C) BFS only
(D) DFS only
Answer: (A) Underfitting
Q#5: Too weak bias can lead to:
(A) Overfitting
(B) Underfitting
(C) BFS only
(D) DFS only
Answer: (A) Overfitting
Q#6: Knowledge can be represented as:
(A) Rules, logic, or probabilistic models
(B) Raw data only
(C) BFS only
(D) DFS only
Answer: (A) Rules, logic, or probabilistic models
Q#7: In decision tree learning, knowledge is captured by:
(A) Tree structure of attribute tests
(B) BFS only
(C) DFS only
(D) Random guesses
Answer: (A) Tree structure of attribute tests
Q#8: Knowledge can guide learning by:
(A) Restricting search space or preferring certain hypotheses
(B) Ignoring all training data
(C) BFS only
(D) DFS only
Answer: (A) Restricting search space or preferring certain hypotheses
Q#9: Domain knowledge in learning helps to:
(A) Improve generalization and reduce required data
(B) Increase randomness
(C) BFS only
(D) DFS only
Answer: (A) Improve generalization and reduce required data
Q#10: Prior knowledge is:
(A) Knowledge available before learning begins
(B) Learned from data only
(C) BFS only
(D) DFS only
Answer: (A) Knowledge available before learning begins
Q#11: Declarative knowledge is:
(A) Knowledge about facts or rules
(B) Knowledge about processes only
(C) BFS only
(D) DFS only
Answer: (A) Knowledge about facts or rules
Q#12: Procedural knowledge is:
(A) Knowledge about how to perform tasks
(B) Facts only
(C) BFS only
(D) DFS only
Answer: (A) Knowledge about how to perform tasks
Q#13: Knowledge can be explicit:
(A) Directly represented and communicated
(B) Hidden in models only
(C) BFS only
(D) DFS only
Answer: (A) Directly represented and communicated
Q#14: Knowledge can be implicit:
(A) Hidden in learned models
(B) Always visible
(C) BFS only
(D) DFS only
Answer: (A) Hidden in learned models
Q#15: A hypothesis space is:
(A) Set of all possible hypotheses considered by a learning algorithm
(B) BFS only
(C) DFS only
(D) Random guesses
Answer: (A) Set of all possible hypotheses considered by a learning algorithm
Q#16: Constraining the hypothesis space helps to:
(A) Reduce overfitting and computational cost
(B) Increase noise
(C) BFS only
(D) DFS only
Answer: (A) Reduce overfitting and computational cost
Q#17: Symbolic knowledge representation uses:
(A) Logical statements, rules, or symbols
(B) Raw numbers only
(C) BFS only
(D) DFS only
Answer: (A) Logical statements, rules, or symbols
Q#18: Subsymbolic knowledge representation uses:
(A) Neural networks or distributed representations
(B) Logical rules only
(C) BFS only
(D) DFS only
Answer: (A) Neural networks or distributed representations
Q#19: Knowledge can guide learning through:
(A) Heuristics, constraints, or preferences
(B) Ignoring data
(C) BFS only
(D) DFS only
Answer: (A) Heuristics, constraints, or preferences
Q#20: Domain theories provide:
(A) Rules about the structure of the problem domain
(B) Raw training data only
(C) BFS only
(D) DFS only
Answer: (A) Rules about the structure of the problem domain
Q#21: Explanation-based learning (EBL) uses:
(A) Domain knowledge to generalize from single examples
(B) Random guesses
(C) BFS only
(D) DFS only
Answer: (A) Domain knowledge to generalize from single examples
Q#22: EBL reduces:
(A) Required training examples
(B) BFS only
(C) DFS only
(D) Random guesses
Answer: (A) Required training examples
Q#23: Knowledge in learning improves:
(A) Accuracy, efficiency, and generalization
(B) Only randomness
(C) BFS only
(D) DFS only
Answer: (A) Accuracy, efficiency, and generalization
Q#24: Prior probabilities in Bayesian learning represent:
(A) Knowledge about likely hypotheses before seeing data
(B) Random noise
(C) BFS only
(D) DFS only
Answer: (A) Knowledge about likely hypotheses before seeing data
Q#25: Posterior probabilities update:
(A) Beliefs after observing training data
(B) Prior knowledge only
(C) BFS only
(D) DFS only
Answer: (A) Beliefs after observing training data
Q#26: Knowledge can prevent:
(A) Overfitting by constraining the model
(B) BFS only
(C) DFS only
(D) Random guesses
Answer: (A) Overfitting by constraining the model
Q#27: Expert knowledge can be encoded as:
(A) Rules, constraints, or heuristics
(B) Random numbers only
(C) BFS only
(D) DFS only
Answer: (A) Rules, constraints, or heuristics
Q#28: Machine learning systems can acquire knowledge from:
(A) Examples, prior theories, or interaction with environment
(B) BFS only
(C) DFS only
(D) Random guesses
Answer: (A) Examples, prior theories, or interaction with environment
Q#29: Knowledge in inductive learning helps to:
(A) Limit hypothesis space for better generalization
(B) Increase search space
(C) BFS only
(D) DFS only
Answer: (A) Limit hypothesis space for better generalization
Q#30: Knowledge-based learning integrates:
(A) Learning algorithms with domain knowledge
(B) BFS only
(C) DFS only
(D) Random guesses
Answer: (A) Learning algorithms with domain knowledge
Q#31: Background knowledge in learning can include:
(A) Ontologies, rules, and constraints
(B) Random noise only
(C) BFS only
(D) DFS only
Answer: (A) Ontologies, rules, and constraints
Q#32: Knowledge representation affects:
(A) Expressiveness and computational efficiency
(B) Random guesses
(C) BFS only
(D) DFS only
Answer: (A) Expressiveness and computational efficiency
Q#33: Declarative knowledge allows:
(A) Reasoning about facts
(B) Procedural execution only
(C) BFS only
(D) DFS only
Answer: (A) Reasoning about facts
Q#34: Procedural knowledge allows:
(A) Performing tasks or operations
(B) Reasoning about facts only
(C) BFS only
(D) DFS only
Answer: (A) Performing tasks or operations
Q#35: Knowledge-based systems combine:
(A) Stored knowledge with inference mechanisms
(B) Raw data only
(C) BFS only
(D) DFS only
Answer: (A) Stored knowledge with inference mechanisms
Q#36: Learning systems can refine knowledge by:
(A) Updating models based on new examples
(B) Ignoring new data
(C) BFS only
(D) DFS only
Answer: (A) Updating models based on new examples
Q#37: Knowledge can be used to generate:
(A) Features, constraints, and heuristics
(B) Random noise
(C) BFS only
(D) DFS only
Answer: (A) Features, constraints, and heuristics
Q#38: Knowledge in learning helps reduce:
(A) Sample complexity
(B) BFS only
(C) DFS only
(D) Random guesses
Answer: (A) Sample complexity
Q#39: Background knowledge can be expressed in:
(A) Logic, probabilistic models, or rules
(B) Random numbers only
(C) BFS only
(D) DFS only
Answer: (A) Logic, probabilistic models, or rules
Q#40: Knowledge can improve:
(A) Learning speed and model interpretability
(B) Only randomness
(C) BFS only
(D) DFS only
Answer: (A) Learning speed and model interpretability
Q#41: Transfer learning uses:
(A) Knowledge learned in one domain to assist learning in another
(B) Random guesses
(C) BFS only
(D) DFS only
Answer: (A) Knowledge learned in one domain to assist learning in another
Q#42: Meta-learning focuses on:
(A) Learning how to learn
(B) Learning a single task only
(C) BFS only
(D) DFS only
Answer: (A) Learning how to learn
Q#43: Knowledge in learning allows:
(A) Faster convergence and improved predictions
(B) BFS only
(C) DFS only
(D) Random guesses
Answer: (A) Faster convergence and improved predictions
Q#44: Feature engineering involves:
(A) Using domain knowledge to create informative input attributes
(B) BFS only
(C) DFS only
(D) Random numbers
Answer: (A) Using domain knowledge to create informative input attributes
Q#45: Knowledge-based bias helps:
(A) Reduce overfitting and improve generalization
(B) Increase noise
(C) BFS only
(D) DFS only
Answer: (A) Reduce overfitting and improve generalization
Q#46: Learning from knowledge complements:
(A) Learning from examples
(B) Random guessing only
(C) BFS only
(D) DFS only
Answer: (A) Learning from examples
Q#47: Domain theories help:
(A) Constrain hypothesis space for inductive learning
(B) Random noise only
(C) BFS only
(D) DFS only
Answer: (A) Constrain hypothesis space for inductive learning
Q#48: Explanations in learning help:
(A) Improve interpretability of learned models
(B) BFS only
(C) DFS only
(D) Random guesses
Answer: (A) Improve interpretability of learned models
Q#49: Knowledge in learning allows algorithms to:
(A) Generalize better from fewer examples
(B) BFS only
(C) DFS only
(D) Memorize all examples only
Answer: (A) Generalize better from fewer examples
Q#50: The main goal of knowledge in learning is:
(A) Guide and constrain learning for better generalization and efficiency
(B) Memorize all data only
(C) BFS only
(D) DFS only
Answer: (A) Guide and constrain learning for better generalization and efficiency