HOMEWORK 5, CSC 190A/290A, 2008 =============================== DUE: Fri Apr 25/08, 5:00pm NOTE: As always, questions should be answered independently by each student (which does not preclude talking about the ideas). And as always, any text taken from elsewhere must be quoted and attributed. Keep your answers as crisp and tightly reasoned as you can. It's better to say a few things clearly and persuasively than to keep writing just to fill the page. For most parts of the 5 questions, a few sentences or a paragraph should suffice. (Perhaps one or two of the questions may justify a longer answer, of a page or so.) ALL WORK SHOULD BE IN ELECTRONIC FORM, BUT PRINTED OUT FOR HANDING IN. No handwritten papers, again -- please. -=-=-=-=-=-=-=-=-=-=- 1. McDermott: Dualism (ch.1-3, 6) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Dualism -- is it even logically coherent? I.e., if there is a world of "spirit" that influences the physical world, then isn't "spirit" by definition physical as well? I.e., doesn't the physical world, by any scientist's definition, include everything that causally influences events in the world? What would it mean to say that the spiritual domain has effects on the physical world, but not causal ones? What could "nonphysical" possibly mean? (Or could we say that the spiritual world is a separate world, with no influence on the physical world? Where would that line of reasoning lead?) [Note: we are not fishing for any particular answers here, just a well-reasoned position.] 2. Computational minds (ch. 5) ~~~~~~~~~~~~~~~~~~~~~~~~~~~ a. Show in detail (based on McDermott's definition of a "computer for a function f") that a device that does nothing at all except that a green light comes on every 5 seconds, lasting for 5 seconds, computes the function f(x) = x^2, b. Varel Akman's criticism of McDermott: At the website http://ndpr.nd.edu/review.cfm?id=1149, philosopher Varol Akman (Bilkent University, Ankara) reviews Mind and Mechanism very favorably, but one criticism he has is that McDermott doesn't connect his perspective with Donald Davidson's: Do I see any shortcomings of Mind and Mechanism? Well, probably just this one. Except for a passing remark (on something else), McDermott does not connect his proposal with the doctrines of Davidson. This is a pity because such a comparison would be promising, even fruitful. As many readers will know, Davidson has worked out an ingenious answer to the puzzle of the mental. Basically, he takes it for granted that the essential properties of matter as described by physicists are the only properties we have. Thus, he subscribes to some form of materialism. However, he thinks that one can be a materialist while also asserting that the mental cannot be reduced to the physical. Assume that you have complete knowledge in front of you of your brain and any relevant neurophysiological systems. According to Davidson, this knowledge cannot constitute knowledge of your beliefs, desires, intentions, etc. This he maintains without really taking a dualist stance, that is, without assuming that your mind has a separate kind of existence. Rather, his point is that our vocabulary for describing the mental does not match the concepts of physics in the right way. For example, he sees the principle of rationality as a most crucial aspect of the mental (especially belief), and holds that this principle has no echo in physical theory. Davidson’s thesis is that the nature of mental phenomena does not permit law-like regularities connecting the mental phenomena with physical events in the brain. Comment -- perhaps the way McDermott might have commented -- on Davidson's contention that in a clear sense, mental phenomena cannot be understood in terms of physical events. Your knowledge of how computers (and robots) work should be helpful here, starting with the physical microcomponents and going "upward" to the level of data structures and programs that implement perception and reasoning techniques. (You might in particular ask yourself whether the behavior of computer programs can be understood in terms of, say, transistor behavior.) Also, you may be helped to some extent by what you have gleaned about the workings of the brain from our readings so far. 3. Knowledge representation and reasoning ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ a. Represent the following sentences in first-order logic (FOL): i. Gort is a large green robot. ii. Klaatu is not Gort's master. [Represent "master" as a 2-place predicate.] iii. Every human is in awe of Gort. [Represent "is in awe of" as a 2-place predicate.] v. Some robots have no master. b. Suppose M = (D,I) is a model of our knowledge representation (which includes the above claims), where as usual D is the domain of individuals and I is the interpretation function. Then i. How is I(Gort) related to D? ii. How is I(master) related to D? iii. In relation to the model D, what does it mean for the wff (well-formed formula) you supplied as your answer to (a.ii) to be true? iv. Suppose that in addition to (a)(i-v), we have human(Klaatu), i.e., Klaatu is a human. Prove formally, using universal instantiation (UI) and modus ponens (MP) that Klaatu is in awe of Gort. c. Are there any difficulties about representing knowing, intending, and other "attitudes" in standard FOL? If so, say what they are; if not, say how we deal with them. 4. Vere & Bickmore's "Homer" ~~~~~~~~~~~~~~~~~~~~~~~~ a. Assume that the semantics of Homer's planning operators is much like the one we are assuming for the Gridworld project, i.e., preconditions are considered true for a particular binding of the parameters if each positive precondition (with that binding) is explicitly present in the representation of the current state, and each negative precondition (with that binding) is absent from the representation of the current state. When these requirements are met, then the assumption is that the operator will have its declared effects (under that binding). Under this semantics, and given the way Homer obtains current-state information, Homer may believe that an operator instance is feasible and will succeed (i.e., achieve its declared effects) when in fact it won't. Why is that? (There are at least two, and quite possibly three, somewhat distinct reasons why this may happen -- you are asked to say what these 2 or 3 reasons are). b. Homer does not have any motivations or preferences of its own. Speculate about what kinds of motivations or preferences might reasonably be given to an enhanced, motivated version of Homer, how this might be done, and how it would affect Homer's behavior, including its interactions with a user. 5. Explicit self-knowledge ~~~~~~~~~~~~~~~~~~~~~~~ a. Compare McCarthy's comments on free will with McDermott's. b. Express the following two claims in the "mental situation calculus" of McCarthy: i. If I know (in a given situation) whether proposition p is true, then either I know that p or I know that not p (in that situation). ii. If I look up a proposition p in an encyclopedia (i.e., to determine whether p is true or false), then as a result of doing so I know whether p is true. c. How does Schubert & Morbini's's approach to the representation of events and situations differ from McCarthy's? d. How does Schubert & Morbini's's approach to the meaning and representation of "knowing" differ from McCarthy's?