Planning Questions and Answers

Q: What is a serialisable goal set?

If no goal's achievement undoes another, the problem decomposes and a planner may work through them in turn, which is why such domains are much easier. The Sussman anomaly is the standard example of a non-serialisable set. Fixed sequences, decomposition into separate problems and importance ordering are different notions. Identifying when decomposition is sound, rather than assuming it, is what separates a correct planner from a plausible one. Checking the units of each quantity catches a wrong formula before the arithmetic begins.

Q: What is the practical role of planning in deployed systems?

Planners schedule spacecraft observations, control autonomous vehicles, route deliveries and sequence manufacturing operations, all domains where the actions and their effects are known and a sequence must be chosen. Question answering, learning and verification are served by other techniques. Identifying the shape of problem a technology suits, rather than its list of successes, is what allows it to be reached for at the right moment. The quantity that varies and the quantity that does not are worth naming before any substitution is attempted.

Q: What characterises a problem suited to a planner?

Planning applies where the dynamics are known and written down, and the question is which sequence of known actions reaches a stated goal. Where the dynamics must be learned, reinforcement learning applies instead, and the other options describe verification and classification. Asking whether the model exists, before choosing between planning and learning, is the question that settles which family of techniques is appropriate.

Q: What is the relationship between planning and reinforcement learning?

Both seek a course of action maximising some objective, and the difference is whether the transition model is given or must be learned from experience. Model based reinforcement learning sits between them by learning a model and then planning with it. Neither contains the other, both handle discrete and continuous cases. Placing two apparently distinct fields on a single axis, according to what is known in advance, makes the relationship between them immediately clear.

Q: What is the frame axiom problem that planning representations avoid?

Writing an axiom for every combination of action and fluent that is not affected would require a quadratic number of statements, most of them trivial, and the planning representation simply stipulates that unmentioned fluents persist. That stipulation is what makes the formalism concise. Reference frames, fluent selection and domain boundaries are different modelling questions. Adopting a convention that removes a quadratic bookkeeping burden is the single most important design decision in the representation.

Q: What is the price of the persistence convention?

The convention requires the effects to be complete, so an action with unknown or context dependent consequences cannot be described without extending the formalism. Conditional effects are a standard such extension. Negation is expressible, states list only true fluents, and conditional effects are supported by most modern languages. Knowing the assumption behind a convenience, and which extensions relax it, is what allows the formalism to be stretched safely.

Q: What is a conditional effect of an action?

Writing that moving a container also moves its contents, when it has any, requires an effect guarded by a condition evaluated in the state where the action occurs. It is how the ramification problem is handled within the formalism. Success conditions, numeric effects and delays are different extensions. Adding guarded effects, rather than writing a separate action for every case, is what keeps a domain description manageable.

Q: Why do conditional effects complicate regression search?

Working backwards must consider both the case where the condition held and the case where it did not, which branches the search. Forward search simply evaluates the condition. Reversibility, action counts and unconditional requirements are different issues. A feature that is cheap in one search direction and expensive in the other is worth noting when choosing between them, since the choice is otherwise made on other grounds entirely. Checking the units of each quantity catches a wrong formula before the arithmetic begins.

Q: What is a precondition of an action?

Preconditions gate applicability, so a planner may apply an action only in states satisfying them, and achieving them becomes a subgoal in backward search. Effects describe what becomes true afterwards. Hopes and goal conditions are different notions. Separating what must hold before from what holds after is the whole structure of an action description, and every planning algorithm is organised around that division.

Q: What is an add effect of an action?

The add list names what becomes true and the delete list what becomes false, and everything unmentioned persists. The delete relaxation ignores the second list, which is what makes the relaxed problem monotonic. Preconditions and unchanged fluents are different categories. Knowing which list a heuristic ignores, and why that makes the relaxed problem easy, connects the representation directly to the heuristics computed from it.

81.

What is a serialisable goal set?

A.One whose goals may be achieved one after another without interference
B.One in which the goals must be achieved in a particular sequence
C.One that can be written as a sequence of separate planning problems
D.One whose goals are listed in the order of their importance

Answer: A

If no goal's achievement undoes another, the problem decomposes and a planner may work through them in turn, which is why such domains are much easier. The Sussman anomaly is the standard example of a non-serialisable set. Fixed sequences, decomposition into separate problems and importance ordering are different notions. Identifying when decomposition is sound, rather than assuming it, is what separates a correct planner from a plausible one. Checking the units of each quantity catches a wrong formula before the arithmetic begins.

Discuss in Forum

82.

What is the practical role of planning in deployed systems?

A.Verifying that a program meets its formal specification
B.Robotics, logistics, spacecraft operations and workflow
C.Answering natural language questions from a knowledge base
D.Learning the effects of actions from observational data

Answer: B

Planners schedule spacecraft observations, control autonomous vehicles, route deliveries and sequence manufacturing operations, all domains where the actions and their effects are known and a sequence must be chosen. Question answering, learning and verification are served by other techniques. Identifying the shape of problem a technology suits, rather than its list of successes, is what allows it to be reached for at the right moment. The quantity that varies and the quantity that does not are worth naming before any substitution is attempted.

Discuss in Forum

83.

What characterises a problem suited to a planner?

A.A requirement to prove a property of a system formally
B.A need to classify observations into known categories
C.Known actions with known effects and a goal to be achieved
D.A large data set from which behaviour must be learned

Answer: C

Planning applies where the dynamics are known and written down, and the question is which sequence of known actions reaches a stated goal. Where the dynamics must be learned, reinforcement learning applies instead, and the other options describe verification and classification. Asking whether the model exists, before choosing between planning and learning, is the question that settles which family of techniques is appropriate.

Discuss in Forum

84.

What is the relationship between planning and reinforcement learning?

A.Reinforcement learning is a special case of classical planning
B.The two solve entirely unrelated classes of problem
C.Planning applies to discrete and learning to continuous problems
D.Planning assumes a known model; learning acquires one by acting

Answer: D

Both seek a course of action maximising some objective, and the difference is whether the transition model is given or must be learned from experience. Model based reinforcement learning sits between them by learning a model and then planning with it. Neither contains the other, both handle discrete and continuous cases. Placing two apparently distinct fields on a single axis, according to what is known in advance, makes the relationship between them immediately clear.

Discuss in Forum

85.

What is the frame axiom problem that planning representations avoid?

A.Stating for every action what it leaves unchanged
B.Deciding which frame of reference to describe the world in
C.Choosing which fluents to include in the state description
D.Determining the boundary of the domain being modelled

Answer: A

Writing an axiom for every combination of action and fluent that is not affected would require a quadratic number of statements, most of them trivial, and the planning representation simply stipulates that unmentioned fluents persist. That stipulation is what makes the formalism concise. Reference frames, fluent selection and domain boundaries are different modelling questions. Adopting a convention that removes a quadratic bookkeeping burden is the single most important design decision in the representation.

Discuss in Forum

86.

What is the price of the persistence convention?

A.It prevents actions from having conditional effects
B.It only works when actions have fully specified effects
C.It makes the representation unable to express any negation
D.It forces every fluent to be listed in every state

Answer: B

The convention requires the effects to be complete, so an action with unknown or context dependent consequences cannot be described without extending the formalism. Conditional effects are a standard such extension. Negation is expressible, states list only true fluents, and conditional effects are supported by most modern languages. Knowing the assumption behind a convenience, and which extensions relax it, is what allows the formalism to be stretched safely.

Discuss in Forum

87.

What is a conditional effect of an action?

A.An effect whose value depends on a numeric computation
B.An effect that takes place after a delay from the action
C.An effect that occurs only when a further condition holds
D.An effect that occurs only if the action succeeds at all

Answer: C

Writing that moving a container also moves its contents, when it has any, requires an effect guarded by a condition evaluated in the state where the action occurs. It is how the ramification problem is handled within the formalism. Success conditions, numeric effects and delays are different extensions. Adding guarded effects, rather than writing a separate action for every case, is what keeps a domain description manageable.

Discuss in Forum

88.

Why do conditional effects complicate regression search?

A.Conditional effects cannot be reversed in any circumstances
B.The number of actions grows with the number of conditions
C.Regression requires every effect to be unconditional
D.Whether an effect occurred depends on the state before the action

Answer: D

Working backwards must consider both the case where the condition held and the case where it did not, which branches the search. Forward search simply evaluates the condition. Reversibility, action counts and unconditional requirements are different issues. A feature that is cheap in one search direction and expensive in the other is worth noting when choosing between them, since the choice is otherwise made on other grounds entirely. Checking the units of each quantity catches a wrong formula before the arithmetic begins.

Discuss in Forum

89.

What is a precondition of an action?

A.A condition that must hold for the action to be applicable
B.A condition that becomes true once the action has completed
C.A condition the planner hopes will hold when the action runs
D.A condition attached to the goal rather than to the action

Answer: A

Preconditions gate applicability, so a planner may apply an action only in states satisfying them, and achieving them becomes a subgoal in backward search. Effects describe what becomes true afterwards. Hopes and goal conditions are different notions. Separating what must hold before from what holds after is the whole structure of an action description, and every planning algorithm is organised around that division.

Discuss in Forum

90.

What is an add effect of an action?

A.A fluent whose value the action leaves unchanged
B.A fluent the action makes true
C.A fluent the action makes false
D.A fluent the action requires to be true beforehand

Answer: B

The add list names what becomes true and the delete list what becomes false, and everything unmentioned persists. The delete relaxation ignores the second list, which is what makes the relaxed problem monotonic. Preconditions and unchanged fluents are different categories. Knowing which list a heuristic ignores, and why that makes the relaxed problem easy, connects the representation directly to the heuristics computed from it.

Discuss in Forum

Loading…

Topics

Other Subjects

Tags: Planning Questions and Answers || Planning MCQ Questions and Answers || Planning GK Questions and Answers || Planning GK MCQ Questions || Planning Multiple Choice Questions and Answers || Planning GK || GK on Planning || Artificial Intelligence Questions and Answers || Artificial Intelligence MCQ Questions and Answers || Artificial Intelligence GK Questions and Answers || GK on Artificial Intelligence

Planning Questions and Answers

Topics

Other Subjects

About Us

Stay informed about all important Dates & Notifications