Find counter arguments by entering NameVs… or …VsName.
The author or concept searched is found in the following 8 entries.
| Disputed term/author/ism | Author |
Entry |
Reference |
|---|---|---|---|
| Backtracking | Norvig | Norvig I 87 Backtracking search/search algorithms/artificial intelligence/Norvig/Russell: Backtracking uses less memory [than depth-first search]. (…) only one successor is generated at a time rather than all successors; each partially expanded node remembers which successor to generate next. In this way, only O(m) memory is needed rather than O(bm) (hwere m is the maximum depth of any node). Backtracking search facilitates yet another memory-saving (and time-saving) trick: the idea of generating a successor by modifying the current state description directly rather than copying it first. This reduces the memory requirements to just one state description and O(m) actions. For this to work, we must be able to undo each modification when we go back to generate the next successor. Norvig I 228 The idea of backtracking search goes back to Golomb and Baumert (1965)(1), and its application to constraint satisfaction is due to Bitner and Reingold (1975)(2), although they trace the basic algorithm back to the 19th century. Bitner and Reingold also introduced the MRV heuristic, which they called the most-constrained-variable heuristic. Brelaz (1979)(3) used the degree heuristic as a tiebreaker after applying the MRV heuristic. The resulting algorithm, despite its simplicity, is still the best method for k-coloring arbitrary graphs. Haralick and Elliot (1980)(4) proposed the least-constraining-value heuristic. Norvig I 229 The basic back jumping method is due to John Gaschnig (1977(5), 1979(6)). Kondrak and van Beek (1997)(7) showed that this algorithm is essentially subsumed by forward checking. Conflict-directed back jumping was devised by Prosser (1993)(8). The most general and powerful form of intelligent backtracking was actually developed very early on by Stallman and Sussman (1977)(9). Their technique of dependency-directed backtracking led to the development of truth maintenance systems (Doyle, 1979)(10) (…). The connection between the two areas is analyzed by de Kleer (1989(11)). For forward chaining, backward chaining: see >Software agents/Norvig. 1. Golomb, S. and Baumert, L. (1965). Backtrack proramming. JACM, 14, 516–524. 2. Bitner, J. R. and Reingold, E. M. (1975). Backtrack programming techniques. CACM, 18(11), 651–656. 3. Brelaz, D. (1979). New methods to color the vertices of a graph. CACM, 22(4), 251–256. 4. Haralick, R. M. and Elliot, G. L. (1980). Increasing tree search efficiency for constraint satisfaction problems. AIJ, 14(3), 263–313. 5. Gaschnig, J. (1977). A general backtrack algorithm that eliminates most redundant tests. In IJCAI-77, p. 457. 6. Gaschnig, J. (1979). Performance measurement and analysis of certain search algorithms. Technical report CMU-CS-79-124, Computer Science Department, Carnegie-Mellon University. 7. Kondrak, G. and van Beek, P. (1997). A theoretical evaluation of selected backtracking algorithms. AIJ, 89, 365–387. 8. Prosser, P. (1993). Hybrid algorithms for constraint satisfaction problems. Computational Intelligence, 9, 268–299. 9. Stallman, R.M. and Sussman, G. J. (1977). Forward reasoning and dependency-directed backtracking in a system for computer-aided circuit analysis. AIJ, 9(2), 135–196 10. Doyle, J. (1979). A truth maintenance system. AIJ, 12(3), 231–272. 11. de Kleer, J. (1989). A comparison of ATMS and CSP techniques. In IJCAI-89, Vol. 1, pp. 290–296. |
Norvig I Peter Norvig Stuart J. Russell Artificial Intelligence: A Modern Approach Upper Saddle River, NJ 2010 |
| Backtracking | Russell | Norvig I 87 Backtracking search/search algorithms/artificial intelligence/Norvig/Russell: Backtracking uses less memory [than depth-first search]. (…) only one successor is generated at a time rather than all successors; each partially expanded node remembers which successor to generate next. In this way, only O(m) memory is needed rather than O(bm). Backtracking search facilitates yet another memory-saving (and time-saving) trick: the idea of generating a successor by modifying the current state description directly rather than copying it first. This reduces the memory requirements to just one state description and O(m) actions. For this to work, we must be able to undo each modification when we go back to generate the next successor. Norvig I 228 The idea of backtracking search goes back to Golomb and Baumert (1965)(1), and its application to constraint satisfaction is due to Bitner and Reingold (1975)(2), although they trace the basic algorithm back to the 19th century. Bitner and Reingold also introduced the MRV heuristic, which they called the most-constrained-variable heuristic. Brelaz (1979)(3) used the degree heuristic as a tiebreaker after applying the MRV heuristic. The resulting algorithm, despite its simplicity, is still the best method for k-coloring arbitrary graphs. Haralick and Elliot (1980)(4) proposed the least-constraining-value heuristic. Norvig I 229 The basic back jumping method is due to John Gaschnig (1977(5), 1979(6)). Kondrak and van Beek (1997)(7) showed that this algorithm is essentially subsumed by forward checking. Conflict-directed back jumping was devised by Prosser (1993)(8). The most general and powerful form of intelligent backtracking was actually developed very early on by Stallman and Sussman (1977)(9). Their technique of dependency-directed backtracking led to the development of truth maintenance systems (Doyle, 1979)(10) (…). The connection between the two areas is analyzed by de Kleer (1989(11)). For forward chaining, backward chaining: see >Software agents/Norvig. 1. Golomb, S. and Baumert, L. (1965). Backtrack proramming. JACM, 14, 516–524. 2. Bitner, J. R. and Reingold, E. M. (1975). Backtrack programming techniques. CACM, 18(11), 651–656. 3. Brelaz, D. (1979). New methods to color the vertices of a graph. CACM, 22(4), 251–256. 4. Haralick, R. M. and Elliot, G. L. (1980). Increasing tree search efficiency for constraint satisfaction problems. AIJ, 14(3), 263–313. 5. Gaschnig, J. (1977). A general backtrack algorithm that eliminates most redundant tests. In IJCAI-77, p. 457. 6. Gaschnig, J. (1979). Performance measurement and analysis of certain search algorithms. Technical report CMU-CS-79-124, Computer Science Department, Carnegie-Mellon University. 7. Kondrak, G. and van Beek, P. (1997). A theoretical evaluation of selected backtracking algorithms. AIJ, 89, 365–387. 8. Prosser, P. (1993). Hybrid algorithms for constraint satisfaction problems. Computational Intelligence, 9, 268–299. 9. Stallman, R.M. and Sussman, G. J. (1977). Forward reasoning and dependency-directed backtracking in a system for computer-aided circuit analysis. AIJ, 9(2), 135–196 10. Doyle, J. (1979). A truth maintenance system. AIJ, 12(3), 231–272. 11. de Kleer, J. (1989). A comparison of ATMS and CSP techniques. In IJCAI-89, Vol. 1, pp. 290–296. |
Russell I B. Russell/A.N. Whitehead Principia Mathematica Frankfurt 1986 Russell II B. Russell The ABC of Relativity, London 1958, 1969 German Edition: Das ABC der Relativitätstheorie Frankfurt 1989 Russell IV B. Russell The Problems of Philosophy, Oxford 1912 German Edition: Probleme der Philosophie Frankfurt 1967 Russell VI B. Russell "The Philosophy of Logical Atomism", in: B. Russell, Logic and KNowledge, ed. R. Ch. Marsh, London 1956, pp. 200-202 German Edition: Die Philosophie des logischen Atomismus In Eigennamen, U. Wolf (Hg) Frankfurt 1993 Russell VII B. Russell On the Nature of Truth and Falsehood, in: B. Russell, The Problems of Philosophy, Oxford 1912 - Dt. "Wahrheit und Falschheit" In Wahrheitstheorien, G. Skirbekk (Hg) Frankfurt 1996 Norvig I Peter Norvig Stuart J. Russell Artificial Intelligence: A Modern Approach Upper Saddle River, NJ 2010 |
| I, Ego, Self | Nozick | II 79 I/use/Nozick: all semantic facts about what the use of "I" refers to, state necessity de dicto, not de re. Cf. >de re, >Semantic facts, >Use, >Mention, >I, Ego, Self, >Reference, >Index words, >Indexicality. II 91 I/synthesis/Nozick: Problem: how do we know that not in any moment a new I is synthesized? Cf. >Apprehension, >Apperception. II 104 I/unit/self/Nozick: unit is not about the act, which could have produced something else - but as a unified whole the I constitutes itself as capable of having other bodily parts or to lose memories (perhaps all). >I/Kant, >I/Fichte, >Memory, >Subject, >Self. II 105 I/self: is projected into the future, as comprising certain stages - after the scheme of the next successor the self-concept will be a listing and weighting of dimensions - but no metric (more Next are possible). >Nearest Successor/Nozick, >Terminology/Nozick, >Similarity Metrics. Nozick: Thesis: we are choosing partially by ourselves. II 112 I/Nozick: physical descriptions exclude me, because they are not reflexive. >Description. II 113 Self/I/Part/Whole/Nozick: a) self as the next successor of each act of synthesis, or b) rather an underlying, enduring self: then rather a whole, less limitations, more unit. >Castaneda: volatile egos. |
No I R. Nozick Philosophical Explanations Oxford 1981 No II R., Nozick The Nature of Rationality 1994 |
| Parts | Nozick | II 99 Part/whole/Nozick: a whole is not equal to the sum: different parts always form another sum, but that may be an equal whole. >Mereology, >Wholes, >Part-of relation, >Mereological sum, >Totality. A body can lose the appendix or get dentures. - Body remains a whole during the time (identical). - The sum is not identical when parts are replaced. >Body, >Identity, >Temporal identity, >Person, >Personal identity, >Continuants. The self (whole) may even lose memories and change goals and dispositions. >Memory, >Actions, >Goals, >Dispositions. Identity of the parts is not sufficient for continuity of the whole: the relations of the parts could be changed. >Relations. The whole is not equal to sum: scheme of the next successor: the n.c. of the sum is the sum ofthe n.c. of the parts. >Next Successor/Nozick. But the next successor of the whole is not the sum of the next successor of the parts (similar for numbers). Later successor: body, but not the sum of the parts. Self: is therefore a whole, not a sum. Whole/criterion: it could also exist if it were made of other parts. II 102 The whole thing must not be a conglomerate. |
No I R. Nozick Philosophical Explanations Oxford 1981 No II R., Nozick The Nature of Rationality 1994 |
| Person | Williams | Nozick II 29f Self/Person/Self-identity/Identity/B. Williams: e.g. two stories together that put us to a mystery: 1st case: one person enters a new body, actually two people exchange their bodies. A-body-person: (now connected with the A-body): has all the memories, knowledge, values, behaviors, etc. of the (earlier, complete) person B - if A could choose which pain should be inflicted after the change, he would choose the A-body for it - because he assumes that he lives in B. 2nd case: someone tells them to endure pain. after that, you will learn that you will undergo a change in your psychological condition - so that you will possess the character of someone else - which frightens you, you don't want to lose your identity and then endure pain. Question: Why did the A-person not have the same fears in the first case? Why is case 1: Transfer of a person to another body and case 2: something that happens to a permanent person? Why does memory play a role in case 1? II 31 Difference 1/2: in 2, B does not acquire the memories of A. Nozick II 29f Identity/Person/Self/B. Williams: e.g. Symmetric case: Outside view: two people swap bodies, A is now in the B-body and decides that B (now in his old A-body) pain should be inflicted instead of him in the new body - inside view (symmetric): You are supposed to get pain inflicted which frightens you, before you should get another character which frightens you even more - you choose the pain for yourself to ward off the loss of the person - other decision, symmetric case. Problem: nothing outside influences A's task and acquisition of a new psyche. Question: how can then two tasks and acquisitions lead to an exchange of bodies? Williams: Thesis: physical identity is a necessary condition for personal identity. II 31 Problem: what happens elsewhere can have no effect on whether A continues to live in the A-body. Williams: Thesis: Physical identity is a necessary condition of personal identity. Nozick II 32 Identity/Person/Self/B. Williams: Principle 1: Identity of something cannot depend on whether there is another thing of any kind. Principle 2: if it is possible that there is another thing that prevents identity, then there is no identity, even if this other thing did not exist. NozickVsWilliams: both principles are wrong. E. g. The Vienna Circle dissolves - several successor groups emerge - then the identity depends on something that happens elsewhere ((s) whether there are several groups). >"closest continuer, . Nozick II 33 Identity/time/next successor/NozikVsWilliams: but dependence on the existence of other things: whether a group can call itself a Vienna circle depends on whether there are other groups in exile - if two things are equally close to the original, there is no next successor (closest continuer). Identity in time: necessary condition: to be next successor. II 35 If God provided causally for identity in time, he would also have to decide how the factors should be weighted. >Ship of Theseus. II 40 It may be that the next successor is not close enough. II 41 Randomly created copy is not a next successor (because of missing causality) - we could have the second one without the first one. II 45 Identity in time/problem: overlapping. >Overlapping, >Identity, >Personal identity, >Continuity, >Change, >Temporal identity. |
EconWilliams I Walter E. Williams Race & Economics: How Much Can Be Blamed on Discrimination? (Hoover Institution Press Publication) Stanford, CA: Hoover Institution Press 2011 WilliamsB I Bernard Williams Ethics and the Limits of Philosophy London 2011 WilliamsM I Michael Williams Problems of Knowledge: A Critical Introduction to Epistemology Oxford 2001 WilliamsM II Michael Williams "Do We (Epistemologists) Need A Theory of Truth?", Philosophical Topics, 14 (1986) pp. 223-42 In Theories of Truth, Paul Horwich Aldershot 1994 No I R. Nozick Philosophical Explanations Oxford 1981 No II R., Nozick The Nature of Rationality 1994 |
| Search Algorithms | Norvig | Norvig I 64 Search Algorithms/Russell/Norvig: uninformed search algorithms [are] algorithms that are given no information about the problem other than its definition. Although some of these algorithms can solve any solvable problem, none of them can do so efficiently. Informed search algorithms, on the other hand, can do quite well given some guidance on where to look for solutions. A. Uninformed search. Norvig I 81 Breadth-first search is a simple strategy in which the root node is expanded first, then all the successors of the root node are expanded next, then their successors, and so on. In general, all the nodes are expanded at a given depth in the search tree before any nodes at the next level are expanded. Norvig I 83 The memory requirements are a bigger problem for breadth-first search than is the execution time. (…) exponential-complexity search problems cannot be solved by uninformed methods for any but the smallest instances. Norvig I 85 Depth-first search always expands the deepest node in the current frontier of the search tree. Both versions are nonoptimal. Norvig I 87 Backtracking search: uses less memory. (…) only one successor is generated at a time rather than all successors; each partially expanded node remembers which successor to generate next. In this way, only O(m) memory is needed rather than O(bm). Backtracking search facilitates yet another memory-saving (and time-saving) trick: the idea of generating a successor by modifying the current state description directly rather than copying it first. This reduces the memory requirements to just one state description and O(m) actions. For this to work, we must be able to undo each modification when we go back to generate the next successor. B. Informed (heuristic) search. Norvig I 92 Best-first search is an instance of the general tree-search or graph-search algorithm in which a node is selected for expansion based on an evaluation function, f(n). The evaluation function is construed as a cost estimate, so the node with the lowest evaluation is expanded first. Greedy best-first search tries to expand the node that is closest to the goal, on the grounds that this is likely to lead to a solution quickly. Thus, it evaluates nodes by using just the heuristic function; that is, f(n) = h(n). Norvig I 108 A general tree search algorithm considers all possible paths to find a solution, whereas a graph search algorithm avoids consideration of redundant paths. Norvig I 120 Online search: here, the agent is faced with a state space that is initially unknown and must be explored. Norvig I 121 Local search algorithms: If the path to the goal does not matter, we might consider a different class of algorithms, ones that do not worry LOCAL SEARCH about paths at all. Local search algorithms operate using a single current node (rather than multiple paths) and generally move only to neighbors of that node. These algorithms have two key advantages: (1) they use very little memory - usually a constant amount; and (2) they can often find reasonable solutions in large or infinite (continuous) state spaces for which systematic algorithms are unsuitable.((s) For Problems: cf. >Local minima (local maxima; for a solution: >Simulated annealing). Norvig I 125 Local beam search: (beam search is a path-based algorithm). The local beam search algorithm keeps track of k states rather than Norvig I 126 just one. It begins with k randomly generated states. At each step, all the successors of all k states are generated. If anyone is a goal, the algorithm halts. Otherwise, it selects the k best successors from the complete list and repeats. >Genetic algorithms. And-Or search problem: see >Terminology/Norvig. Norvig I 147 Online search: >Online search/Norvig. Norvig I 154 Literature for local search: (Newton, 1671(1); Raphson, 1690(2)) can be seen as a very efficient local search method for continuous spaces in which gradient information is available. Brent (1973)(3) is a classic reference for optimization algorithms that do not require such information. Beam search, which we have presented as a local search algorithm, originated as a bounded-width variant of dynamic programming for speech recognition in the HARPY system (Lowerre, 1976)(4). A related algorithm is analyzed in depth by Pearl (1984(5)). The topic of local search was reinvigorated in the early 1990s by surprisingly good results for large constraint-satisfaction problems such as n-queens (Minton et al., 1992)(6) and logical reasoning (Selman et al., 1992)(7) and by the incorporation of randomness, multiple simultaneous searches, and other improvements. Tabu serarch: a variant of hill climbing called tabu search has gained popularity (Glover and Laguna, 1997)(8). This algorithm maintains a tabu list of k previously visited states that cannot be revisited; as well as improving efficiency when searching graphs, this list can allow the algorithm to escape from some local minima. Stage algorithm: Another useful improvement on hill climbing is the stage algorithm (Boyan and Moore, 1998)(9). The idea is to use the local maxima found by random-restart hill climbing to get an idea of the overall shape of the landscape. >Constraint Satisfaction Problems/Norvig. Norvig I 227 Constraint satisfaction problems (CSPs) represent a state with a set of variable/value pairs and represent the conditions for a solution by a set of constraints on the variables. Many important real-world problems can be described as CSPs. A number of inference techniques use the constraints to infer which variable/value pairs are consistent and which are not. These include node, arc, path, and k-consistency. Backtracking search, a form of depth-first search, is commonly used for solving CSPs. Inference can be interwoven with search. The minimum-remaining-values and degree heuristics are domain-independent methods for deciding which variable to choose next in a backtracking search. The least constraining- value heuristic helps in deciding which value to try first for a given variable. Backtracking occurs when no legal assignment can be found for a variable. Conflict-directed backjumping backtracks directly to the source of the problem. Local search using the min-conflicts heuristic has also been applied to constraint satisfaction problems with great success. For forward chaining, backward chaining: see >Software agents/Norvig. 1. Newton, I. (1664–1671). Methodus fluxionum et serierum infinitarum. Unpublished notes 2. Raphson, J. (1690). Analysis aequationum universalis. Apud Abelem Swalle, London. 3. Brent, R. P. (1973). Algorithms for minimization without derivatives. Prentice-Hall 4. Lowerre, B. T. (1976). The HARPY Speech Recognition System. Ph.D. thesis, Computer Science Department, Carnegie-Mellon University. 5. Pearl, J. (1984). Heuristics: Intelligent Search Strategies for Computer Problem Solving. Addison- Wesley. 6. Minton, S., Johnston, M. D., Philips, A. B., and Laird, P. (1992). Minimizing conflicts: A heuristic repair method for constraint satisfaction and scheduling problems. AIJ, 58(1–3), 161–205. 7. Selman, B., Levesque, H. J., and Mitchell, D. (1992). A new method for solving hard satisfiability problems. In AAAI-92, pp. 440–446. 8. Glover, F. and Laguna, M. (Eds.). (1997). Tabu search. Kluwer 9. Boyan, J. A. and Moore, A. W. (1998). Learning evaluation functions for global optimization and Boolean satisfiability. In AAAI-98 |
Norvig I Peter Norvig Stuart J. Russell Artificial Intelligence: A Modern Approach Upper Saddle River, NJ 2010 |
| Search Algorithms | Russell | Norvig I 64 Search Algorithms/Russell/Norvig: uninformed search algorithms [are] algorithms that are given no information about the problem other than its definition. Although some of these algorithms can solve any solvable problem, none of them can do so efficiently. Informed search algorithms, on the other hand, can do quite well given some guidance on where to look for solutions. A. Uninformed search. Norvig I 81 Breadth-first search is a simple strategy in which the root node is expanded first, then all the successors of the root node are expanded next, then their successors, and so on. In general, all the nodes are expanded at a given depth in the search tree before any nodes at the next level are expanded. Norvig I 83 The memory requirements are a bigger problem for breadth-first search than is the execution time. (…) exponential-complexity search problems cannot be solved by uninformed methods for any but the smallest instances. Norvig I 85 Depth-first search always expands the deepest node in the current frontier of the search tree. Both versions are nonoptimal. Norvig I 87 Backtracking search: uses less memory. (…) only one successor is generated at a time rather than all successors; each partially expanded node remembers which successor to generate next. In this way, only O(m) memory is needed rather than O(bm). Backtracking search facilitates yet another memory-saving (and time-saving) trick: the idea of generating a successor by modifying the current state description directly rather than copying it first. This reduces the memory requirements to just one state description and O(m) actions. For this to work, we must be able to undo each modification when we go back to generate the next successor. B. Informed (heuristic) search. Norvig I 92 Best-first search is an instance of the general tree-search or graph-search algorithm in which a node is selected for expansion based on an evaluation function, f(n). The evaluation function is construed as a cost estimate, so the node with the lowest evaluation is expanded first. Greedy best-first search tries to expand the node that is closest to the goal, on the grounds that this is likely to lead to a solution quickly. Thus, it evaluates nodes by using just the heuristic function; that is, f(n) = h(n). Norvig I 108 A general tree search algorithm considers all possible paths to find a solution, whereas a graph search algorithm avoids consideration of redundant paths. Norvig I 120 Online search: here, the agent is faced with a state space that is initially unknown and must be explored. Norvig I 121 Local search algorithms: If the path to the goal does not matter, we might consider a different class of algorithms, ones that do not worry LOCAL SEARCH about paths at all. Local search algorithms operate using a single current node (rather than multiple paths) and generally move only to neighbors of that node. These algorithms have two key advantages: (1) they use very little memory - usually a constant amount; and (2) they can often find reasonable solutions in large or infinite (continuous) state spaces for which systematic algorithms are unsuitable.((s) For Problems: cf. >Local minima (local maxima; for a solution: >Simulated annealing). Norvig I 125 Local beam search: (beam search is a path-based algorithm). The local beam search algorithm keeps track of k states rather than Norvig I 126 just one. It begins with k randomly generated states. At each step, all the successors of all k states are generated. If anyone is a goal, the algorithm halts. Otherwise, it selects the k best successors from the complete list and repeats. >Genetic algorithms. And-Or search problem: see >Terminology/Norvig. Norvig I 147 Online search: >Online search/Norvig. Norvig I 154 Literature for local search: (Newton, 1671(1); Raphson, 1690(2)) can be seen as a very efficient local search method for continuous spaces in which gradient information is available. Brent (1973)(3) is a classic reference for optimization algorithms that do not require such information. Beam search, which we have presented as a local search algorithm, originated as a bounded-width variant of dynamic programming for speech recognition in the HARPY system (Lowerre, 1976)(4). A related algorithm is analyzed in depth by Pearl (1984(5)). The topic of local search was reinvigorated in the early 1990s by surprisingly good results for large constraint-satisfaction problems such as n-queens (Minton et al., 1992)(6) and logical reasoning (Selman et al., 1992)(7) and by the incorporation of randomness, multiple simultaneous searches, and other improvements. Tabu serarch: a variant of hill climbing called tabu search has gained popularity (Glover and Laguna, 1997)(8). This algorithm maintains a tabu list of k previously visited states that cannot be revisited; as well as improving efficiency when searching graphs, this list can allow the algorithm to escape from some local minima. Stage algorithm: Another useful improvement on hill climbing is the stage algorithm (Boyan and Moore, 1998)(9). The idea is to use the local maxima found by random-restart hill climbing to get an idea of the overall shape of the landscape. >Constraint Satisfaction Problems/Norvig. Norvig I 227 Constraint satisfaction problems (CSPs) represent a state with a set of variable/value pairs and represent the conditions for a solution by a set of constraints on the variables. Many important real-world problems can be described as CSPs. A number of inference techniques use the constraints to infer which variable/value pairs are consistent and which are not. These include node, arc, path, and k-consistency. Backtracking search, a form of depth-first search, is commonly used for solving CSPs. Inference can be interwoven with search. The minimum-remaining-values and degree heuristics are domain-independent methods for deciding which variable to choose next in a backtracking search. The least constraining- value heuristic helps in deciding which value to try first for a given variable. Backtracking occurs when no legal assignment can be found for a variable. Conflict-directed backjumping backtracks directly to the source of the problem. Local search using the min-conflicts heuristic has also been applied to constraint satisfaction problems with great success. For forward chaining, backward chaining see >Software agents/Norvig. 1. Newton, I. (1664–1671). Methodus fluxionum et serierum infinitarum. Unpublished notes 2. Raphson, J. (1690). Analysis aequationum universalis. Apud Abelem Swalle, London. 3. Brent, R. P. (1973). Algorithms for minimization without derivatives. Prentice-Hall 4. Lowerre, B. T. (1976). The HARPY Speech Recognition System. Ph.D. thesis, Computer Science Department, Carnegie-Mellon University. 5. Pearl, J. (1984). Heuristics: Intelligent Search Strategies for Computer Problem Solving. Addison- Wesley. 6. Minton, S., Johnston, M. D., Philips, A. B., and Laird, P. (1992). Minimizing conflicts: A heuristic repair method for constraint satisfaction and scheduling problems. AIJ, 58(1–3), 161–205. 7. Selman, B., Levesque, H. J., and Mitchell, D. (1992). A new method for solving hard satisfiability problems. In AAAI-92, pp. 440–446. 8. Glover, F. and Laguna, M. (Eds.). (1997). Tabu search. Kluwer 9. Boyan, J. A. and Moore, A. W. (1998). Learning evaluation functions for global optimization and Boolean satisfiability. In AAAI-98 |
Russell I B. Russell/A.N. Whitehead Principia Mathematica Frankfurt 1986 Russell II B. Russell The ABC of Relativity, London 1958, 1969 German Edition: Das ABC der Relativitätstheorie Frankfurt 1989 Russell IV B. Russell The Problems of Philosophy, Oxford 1912 German Edition: Probleme der Philosophie Frankfurt 1967 Russell VI B. Russell "The Philosophy of Logical Atomism", in: B. Russell, Logic and KNowledge, ed. R. Ch. Marsh, London 1956, pp. 200-202 German Edition: Die Philosophie des logischen Atomismus In Eigennamen, U. Wolf (Hg) Frankfurt 1993 Russell VII B. Russell On the Nature of Truth and Falsehood, in: B. Russell, The Problems of Philosophy, Oxford 1912 - Dt. "Wahrheit und Falschheit" In Wahrheitstheorien, G. Skirbekk (Hg) Frankfurt 1996 Norvig I Peter Norvig Stuart J. Russell Artificial Intelligence: A Modern Approach Upper Saddle River, NJ 2010 |
| Self | Nozick | II 61 Identity/time/self/I/Nozick: because it is highly conceptual, a scheme appears to be necessary. >Identity conditions, cf. >Qua objects. This scheme should choose between near and immediate successors, etc. >Next successor/Nozick, >Similarity metrics, >Similarity. But I do not need the scheme to find out my goals, but to find out whose goals are mine. >Self-knowledge, >Self-identification. Problem: who applies actually the scheme? II 78 Definition I/self/some authors: to be an I or to be self, means to have the ability of reflexive self-reference. >I, Ego, >Self, >I/Nozick. NozickVs: 1. This ability must just have existed sometime 2. beings: from the fact that I have this ability, it does not follow that it is essential. In addition: the reflexive self-reference gives me access to the self, but that does not guarantee that it is part of my nature as self. >Self-reference. |
No I R. Nozick Philosophical Explanations Oxford 1981 No II R., Nozick The Nature of Rationality 1994 |
| |||
| Disputed term/author/ism | Author Vs Author |
Entry |
Reference |
|---|---|---|---|
| Williams, B. | Nozick Vs Williams, B. | II 29 Self/Person/Self-Identity/Identity/B.Williams: E.g. two stories that put together present us with a mystery: Case 1: a person enters a new body, or rather two persons exchange their bodies. Two persons, A and B enter a machine A body person: (now connected to the body A): has all the memories, all the knowledge, values, behaviors, etc. of the (former, complete) person B. In the body A is now the "vector product" of this B material with the physical boundaries of body A. Similarly, all the other way round for B. The situation is symmetrical. II 29/30 If A were to decide (after substitutions) now, which severe pain should be inflicted by the two bodies, then A would select the A body for it! Because he believes that he himself inhabits the B body. Case 2: Imagine someone tells them that they are to endure terrible pain. That frightens them. Next, they get the information that they will undergo an enormous change in their psychological constitution, perhaps to the extent that they will have exactly the same character, the memories and behaviors of someone else, who is currently alive. That will scare them even more. They do not want to lose their identity and suffer pain afterwards. Williams: question: why had person A not exactly the same concerns when she heard the first story, as in Case 2? What makes the first story a story about the transfer of a person to a different body and not a story about something that happens to a person who remains who they are? How can the difference consist in that in the first case, in addition to what happens to body A, II 31 also A's memories and mind end or are newly created in body B? Problem: what happens anywhere else can have no effect on whether A continues to live in body A. If this happens to a body, it is a psychological task and the acquisition of a new psyche. Question: how can two tasks and the acquisition of new memories and values result in the exchange of two bodies? Body A / B Body 1) Situation acquires memories + character of B/acquires memories + character of A 2) Situation acquires memories + character of B/keeps memories + character or perhaps entirely new Two principles should explain this: Principle 1/Williams: If x at t1 is the same individual as y after t2, then this can only depend on facts about x, y and the relations between them. No facts about any other existing thing are relevant. That entails: Principle 2/Williams: if y at t2 (is part of the same continuous particular like) x at t1, by virtue of a relation R to x at t1, then there could be another additional thing z at t2 that also (together with y) stands in R to x at t1. If this additional thing z at t2 exists, then neither z nor y would be identical to x. If this z could potentially exist now, although it does currently not exist, then y at t2 is not identical with y at t1, at least not by virtue of relation R! ((s) If there is a relation R that allows identity at a later time, then several things can "benefit" from that and then the identity (which must be unique) would be destroyed. This is true even if the existence of a second thing is merely possible.) II 32 Self/Identity/Person/Williams: Williams had formulated these two principles in three earlier publications to support his thesis: Physical identity is a necessary condition of personal identity. Otherwise it would be possible to imagine that e.g. a person enters a machine, disappears and appears again in another machine at a distance without having crossed the space between them. Or: E.g. There could be a third machine on the other side from which an also (qualitatively) different identical being emerges. Neither would be the original person who had entered the machine in the middle. Now, if in this case of double materialization the original person is not identical with either of the two later persons, so not even in the first case, where only one person appears in a different place. Williams: the mere possibility that someone appears intermittently in another place is sufficient to show that he himself cannot be the same person without doubling. 1) Principle: Identity of something cannot depend on whether there is another thing of some sort. 2) Principle: if it is possible that there was another thing that prevented identity, then there is no identity, even if this other thing did not exist! ((s) The first follows from the second here). NozickVsWilliams: both principles are wrong. 1) (without personal identity): E.g. the Vienna Circle was expelled from Vienna by the Nazis, one member, Reichenbach, came to Istanbul. Suppose there were 20 members of the circle, three of which went to Istanbul and continued to meet. In 1943, they hear that the others are dead. Now they are the Vienna Circle which meets in Istanbul. ((s) ArmstrongVs/ChisholmVs: a local property is not a property.) In 1945, they learn that 9 other members continued to meet in America and further developed the same philosophical program. Nozick: then the group in America is the Vienna Circle, the one in Istanbul is just the offshoot. Nozick: how is that possible? Either the group in Istanbul is the Vienna Circle or it is not. How can this be influenced by something that takes place elsewhere? ((s) Because subsets play a role here, which do not play a role, e.g. in personal identity. Analogue would have been to assume that some of the psychological characteristics are kept during the body changes). II 33 Nozick: E.g. would it not be clear that if the 9 others had survived living underground in Vienna, this would show that the Istanbul group is not the Vienna Circle? So the First Principle (Williams) cannot be applied here: it is not plausible to say that if the group of three in Istanbul is the same entity as the original Vienna Circle, that this can only depend on relations between the two ... Nozick: ...and not on whether anything else exists. Def "Next Successor"/Closest Continuer/Nozick: Solution: The Istanbul group is the next successor. Namely so if no other group exists. But if the group in America exists, it is the next successor. Which one constitutes the Vienna Circle depends (unlike Williams) on the existence of other things. Being something later means being the next successor. ((s) and being able to be called later then depends on the amount of shared properties). E.g. How many other groups of the Vienna Circle are there in exile? ("Scheme"). Identity in Time:/Nozick: it is no problem for something to replace its parts and to keep the identity. E.g. Ship of Theseus/Nozick: 2nd ship made of collection of discarded parts from the old ship: two originals? (Was already known in this form in antiquity). Next Successor: helps to structure the problem, but not solve it. Because the scheme does not say of itself, which dimension of weighted sum of dimensions determine the proximity. Two possibilities: a) spatio-temporal continuity b) continuity of the parts. If both are weighted equally, there is a stalemate. II 34 Neither of them is the next successor. And therefore none is the original. But even if one originally existed without the other, it would be the original as next successor. Perhaps the situation is not a stalemate, but an unclear weighting, the concepts may not be sharp enough to rank all possible combinations. Personal Identity/Nozick: this is different, especially when it comes to ourselves: here we are not ready, that it is a question of decision of the stipulation. Ship of Theseus/NozickVsWilliams: external facts about external things do matter: when we first hear the story, we are not in doubt, only once the variant with the second, reconstructed ship comes into play. Next Successor/Nozick: necessary condition for identity: something at t2 is not the same entity as x at t1 if it is not x's next successor. If two things are equally close, none of them is the next successor. Something can be the next successor of x without being close enough to x to be x itself! If the view of the next successor is correct, then our judgments about identity reflect weights of dimensions. Form of thought: reversal: we can conversely use these judgments to discover these dimensions. II 35 A property may be a factor for identity without being a necessary condition for it. Physical identity can also be an important factor. If something is the next successor, it does not mean that his properties are qualitatively the same as those of x, or are similar to them! Rather, they arise from the properties of x. They are definitely causally caused! Spatio-Temporal Continuity/Nozick: cannot be explained merely as a film without gaps. Counter-example: The replacement with another thing would not destroy the continuity of the film! Causal Relation/Next Successor: the causal relation does not need to involve temporal continuity! E.g. every single thing only possessed a flickering existence (like messages through the telephone). If this applies to all things, it is the best kind of continuity. NozickVsWilliams: but if you find that some things are not subject to the flickering of their existence, then you will no longer talk of other things as the best realizations of continuously existing things. Dependency of identity on other things! Theology/God/Identity/Nozick: Problem: if the causal component is required, and suppose God keeps everything in continuous existence, closing all causal connections in the process: how does God then distinguish the preservation of an old thing in continuity from the production of a new, qualitatively identical thing without interrupting a "movie"? II 36 Temporal Continuity/NozickVsWilliams: how much temporal continuity is necessary for a continuous object depends on how closely things are continuously related elsewhere. Psychology/Continuity/Identity/Nozick: experiments with objects which emerge (again) more or less changed after a time behind a screen. |
No I R. Nozick Philosophical Explanations Oxford 1981 No II R., Nozick The Nature of Rationality 1994 |
| |||