G. Ramalingam. Bounded Incremental Computation, volume 1089 of Lec. Notes in Comp. Sci. Springer-Verlag, 1996.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....of the arrays a, b, and d of Equations 1, 2, and 3 can be maintained with 7 Figure 4: Closing a Warehouse. priority queues in O(m log n) We now show that updating the entries in Equations 4, 5, and 6 takes time linear in the number of updated values. This type of analysis was suggested in [13], since it captures the essence of many incremental algorithms maintaining an output under changes. It is particularly appropriate to analyze local search algorithms, since a move does not change the state much in general. Updating g does not induce any asymptotic cost. Indeed, the update rule ....

G. Ramalingam. Bounded Incremental Computation. PhD thesis, University of Wisconsin-Madison, 1993. 14

....of as a generalization of the certificate chain discovery algorithm. The general strategy is as follows: the set of labeled SPKI SDSI certificates is first translated to a weighted pushdown system. After the translation, the answer is obtained by solving a generalized shortest path problem [27, 46, 34]. The main contributions of the work reported in the paper are as follows: The GAP framework. We define the generalized authorization problem and show how versions of several types of security issues related to authorization can be handled in the GAP framework. An efficient algorithm for ....

....characterizes the sequences of transitions that can be made by a pushdown system for C. We then turn to weighted pushdown systems and the GPR problem. We use the language characterizations of transition sequences, together with previously known results on a certain kind of grammar problem [46, 34] to obtain a solution to the GPR problem. However, the solution based on grammars is somewhat inefficient; to improve the performance, we specialize the computation to our case, ending up with an algorithm for creating an annotated automaton that is quite similar to the pre # algorithm from ....

[Article contains additional citation context not shown here]

G. Ramalingam. Bounded Incremental Computation, volume 1089 of Lec. Notes in Comp. Sci. Springer-Verlag, 1996.

....here uses sequential models of computation, such as the RAM (random access machine) model. Ramalingam and Reps have described a class of dynamic problems by bounding the time allowed for an update computation as a function of the combined change in the input and in the output data structure [36, 37]. Their work investigates problems with a data structure as input and a data structure as output. The output is typically an annotated version of the input data structure. In this model, they obtain linear and thus optimal bounds on the complexity of certain dynamic computations. 86 There is a ....

Ramalingam, G. Bounded incremental computation, vol. 1089 of Lecture Notes in Computer Science. Springer-Verlag Inc., New York, NY, USA, 1996.

....in programming languages. It integrates one way constraints pioneered in the Sketchpad and ThingLab object oriented systems [24, 4] and generalizes them to accommodate nite di erencing techniques on algebraic and set expressions [19, 29] It also encapsulates ecient incremental graph algorithms [21, 1], and uses polymorphism heavily to obtain its compositional nature. Observe also that the resulting architecture has some avor of aspect oriented programming [13] since constraints represent and maintain properties across a wide range of objects. Finally, note also that many of the concepts ....

G. Ramalingam. Bounded Incremental Computation. PhD thesis, University of Wisconsin-Madison, 1993.

....elements in set Moves whose gains are equal to bestGain. Localizer applies incremental algorithms to maintain these invariants under changes to their variables. These algorithms are, in general, optimal in the size of the input and output changes, a model of computational complexity proposed in [28, 29] to analyze incremental algorithms. For instance, when an element of a is changed in the above invariant, Localizer updates variable sum in constant time. Similarly, the set variable BestMoves is updated in constant time when 3 elements are added or deleted from set Moves or when an element in ....

G. Ramalingam. Bounded Incremental Computation. PhD thesis, University of WisconsinMadison, 1993.

....problem is interesting on its own and finds many important applications, including network optimization, document formatting, routing in communication systems, robotics. For a comprehensive review of the application settings for the static and dynamic shortest paths problem, we refer to [1] and [15], respectively. Several theoretical results have been provided in the literature for the dynamic maintenance of shortest paths in graphs with positive arc weights (see, Partially supported by the IST Programme of the EU under contract n. IST 199914186 (ALCOM FT) e.g. 7, 9, 15, 16] We are ....

....we refer to [1] and [15] respectively. Several theoretical results have been provided in the literature for the dynamic maintenance of shortest paths in graphs with positive arc weights (see, Partially supported by the IST Programme of the EU under contract n. IST 199914186 (ALCOM FT) e.g. [7, 9, 15, 16]) We are aware of few efficient fully dynamic solutions for updating shortest paths in general digraphs with arbitrary (positive and non positive) arc weights [10, 16] Recently, an equally important research effort has been done in the field of algorithm engineering, aiming at bridging the gap ....

[Article contains additional citation context not shown here]

G. Ramalingam. Bounded incremental computation. Lect. Notes in Comp. Sc. 1089, 1996.

....elements in set Moves whose gains are equal to bestGain. Localizer applies incremental algorithms to maintain these invariants under changes to their variables. These algorithms are, in general, optimal in the size of the input and output changes, a model of computational complexity proposed in [30, 31] to analyze incremental algorithms. For 3 instance, when an element of a is changed in the above invariant, Localizer updates variable sum in constant time. Similarly, the set variable BestMoves is updated in constant time when elements are added or deleted from set Moves or when an element in ....

G. Ramalingam. Bounded Incremental Computation. PhD thesis, University of WisconsinMadison, 1993.

.... e Sistemistica, Universit a di Roma La Sapienza , Via Salaria 113 00198 Roma, Italy, ffrigioni,ioffreda,nanni,pasqualog dis.uniroma1.it 1 1 Introduction A lot of efforts have been done in the last years in order to devise efficient algorithms for dynamic graph problems (e.g. see [6, 9, 13, 14, 15, 16, 18, 20, 23, 24, 25, 26, 30, 31, 32]) motivated by theoretical as well as practical applications. In the literature, the most used dynamic model is the following: we are given a graph G and we want to answer queries on a property P of G, while the graph is changing due to insertions and deletions of edges. For instance, if the ....

.... This problem is interesting on its own and finds many important applications, including network optimization, document formatting, routing in communication systems, robotics [7] For a comprehensive review of the application settings for the static and dynamic shortest paths problem, we refer to [1, 30]. Several theoretical results have been provided in the literature for the dynamic maintenance of shortest paths in graphs, but, to the best of our knowledge, nothing is known from the experimental point of view. In this paper we provide the first experimental study of dynamic algorithms for the ....

[Article contains additional citation context not shown here]

G. Ramalingam. Bounded Incremental Computation. Lect. Notes in Comp. Sci., 1089, 1996.

.... Delta k Delta log n) We remark that the notion of k bounded accounting function is useful only to bound the running times, but does not affect the behavior of our algorithms. B) The analysis of dynamic algorithms using the output complexity model has been introduced by Ramalingam and Reps in [15, 17] and it has been subsequently modified by the authors of this paper in [10] Other authors used similar concepts in [2] In [15, 17] Ramalingam and Reps propose also the only dynamic solution for shortest paths on digraphs with arbitrary arc weights known in the literature. In this solution they ....

....does not affect the behavior of our algorithms. B) The analysis of dynamic algorithms using the output complexity model has been introduced by Ramalingam and Reps in [15, 17] and it has been subsequently modified by the authors of this paper in [10] Other authors used similar concepts in [2] In [15, 17] Ramalingam and Reps propose also the only dynamic solution for shortest paths on digraphs with arbitrary arc weights known in the literature. In this solution they assume the digraph has no negative length cycles before and after any input change. In addition, they do not deal with zero length ....

G. Ramalingam. Bounded incremental computation. Lecture Notes in Computer Science 1089, 1996.

....the formatter (T E X, L A T E X, etc. will not have to reformat the whole document to obtain the new layout. We are also motivated by previous works on incremental attribute evaluation in language based editors [76] augmented parsers for supporting incrementality [25] incremental computation [74] and expanded querying power in constraint logic programming languages [61, 94] Parallel Implementation: One of the main advantages of declarative programming languages is that they lend themselves to efficient parallel execution. In logic programming, one usually has two kinds of parallelism: ....

G. Ramalingam. Bounded Incremental Computation. PhD thesis, University of Wisconsin, Madison, August 1993.

....shown that in the worst case our algorithm possesses the same CHAPTER 4. INCREMENTAL MODEL CHECKING 70 asymptotic running time as the original non incremental algorithm. We would like to know whether it is possible to gain an asymptotic speedup in the incremental case. The results put forward in [65] suggest that there are limits to what can be achieved. There, a classification of incremental problems with respect to the size of input change is presented. We show that our problem falls into the category of unbounded problems, i.e. the running time of an incremental update cannot be expressed ....

....is presented. We show that our problem falls into the category of unbounded problems, i.e. the running time of an incremental update cannot be expressed solely in terms of the size of input change. Since we do not have values associated with edges, we slightly simplify the original definition from [65] to depend only on the vertices of the graph. The size of the input change is defined in terms of the neighbors of vertices, immediately affected by an update. If K is a subset of vertices of a graph G, then the set of neighbors of order i is defined inductively as N i G (K) NG (N i Gamma1 G ....

[Article contains additional citation context not shown here]

G. Ramalingam. Bounded Incremental Computation. PhD thesis, Computer Sciences Dept., University of Wisconsin-Madison, 1993.

....DAG) differing only in that addition is not an idempotent meet operator. In fact, the problem of calculating numPaths values is an example of an algebraic path problem on a DAG. For a more general discussion of the relationship between algebraic path problems and dataflow analysis problems, see [13]. We now review the appropriate part of Sharir and Pnueli s work [17] with some rephrasing of their work to describe backwards dataflow analysis problems instead of forwards dataflow analysis problems. We then show how their OE functions are related to our functions. In Sharir and Pnueli s ....

G. Ramalingam. Bounded Incremental Computation. Springer-Verlag, 1996.

....existence of the value can be performed if updates may be incorrect; however, it would require maintaining the multiset and would take logarithmic time, which may change the incremental cost of the aggregate computation. 2. 2 Incremental Aggregate Functions The following definitions are based on [4, 12]. Definition 2.2 (Incremental Aggregate Algorithm) Let G: M(D) D 0 be an aggregate function, with domain M(D) being the set of problem inputs, and range D 0 being the set of problem outputs. The size of a problem instance, i.e. the size of a multiset p 2 M(D) is denoted by n. Let U ....

G. Ramalingam. Bounded Incremental Computation. PhD thesis, University of Wisconsin, Madison, Aug. 1993. Technical Report #1172.

....than theirs. 4. Since the two algorithms are fundamentally different, it is difficult to make precise statements on their timing comparisons without substantial tests on real programs. In particular, it is generally very difficult to analytically compare incremental algorithms [Mar89, Ram93] Therefore, we only make some qualitative observations. Our algorithm, for the insertion case, begins by computing precisely the set of affected nodes, at the expense of visiting arguably more nodes compared to the RR algorithm. This is true even for the deletion case) On the other hand, the ....

G. Ramalingam. Bounded Incremental Computation. PhD thesis, University of Wisconsin-Madison, Madison, Wisconsin, August 1993.

....where the user is expected to see any change, explicit updates must be carried out after that an input modification is specified. In these situations a different measure of performance may be considered, beside the classical worstcase and amortized analysis [21] Ramalingam and Reps in [18, 19] propose to measure the cost of a dynamic algorithm in terms of output complexity; a similar notion was also used by other authors (see, e.g. 2] An input modification is defined as a set of edge operations (insertions, deletions, or cost updates) transforming a graph G = V; E) to a new ....

....O(k log n) amortized time per vertex update. Our solution is within a logarithmic factor far away from an optimal solution with explicit updates. For the mentioned classes of graphs and updates (note that Ramalingam and Reps solution is fully dynamic) our results improve the time bounds given in [18, 19], expressed in terms of parameter jjffijj. In fact jffij jjffijj n Delta jffij; furthermore there are cases in which the upper bound is tight. In the particular case of planar graph, the optimal offline solution by Klein et al. 13] requiring O(n) time) may be better than our dynamic ....

[Article contains additional citation context not shown here]

G. Ramalingam, Bounded Incremental Computation, Technical Report 1172, Computer Sciences Department, University of Wisconsin, Madison, WI (1993).

....its arguments and returns True , if there is a path connecting u and v in the current graph. The field of dynamic graph algorithms has been a blossoming field of research in the last years [4, 9, 11, 13, 14, 15, 17, 19, 31, 33] motivated by theoretical and practical questions (see for instance [29]) However, despite this blend of theoretical and practical interests, we are aware of no implementations and experimental studies in this field. In this paper, we aim at bridging this gap by studying the practical properties of theoretically interesting dynamic graph algorithms that were ....

G. Ramalingam, "Bounded incremental computation", Ph.D. Thesis, Department of Computer Science, University of Wisconsin at Madison, August 1993.

....existence of the value can be performed if updates may be incorrect; however, it would require maintaining the multiset and would take logarithmic time, which may change the incremental cost of the aggregate computation. 2. 2 Incremental Aggregate Functions The following definitions are based on [4, 12]. Definition 2.2 (Incremental Aggregate Algorithm) Let G: M(D) D 0 be an aggregate function, with domain M(D) being the set of problem inputs, and range D 0 being the set of problem outputs. The size of a problem instance, i.e. the size of a multiset p 2 M(D) is denoted by n. Let U ....

G. Ramalingam. Bounded Incremental Computation. PhD thesis, University of Wisconsin, Madison, Aug. 1993. Technical Report #1172.

....This is to be expected for it is easy to construct an example in which the value of every variable changes as the result of adding a transition to the LTS. Thus, every node of the product graph must be visited during the incremental run. In fact we prove, via a reduction from SS REACHABILITY (see [Ram93] that model checking is an unbounded problem, meaning that the running time of an incremental update cannot, in general, be expressed solely in terms of Delta. In the best case, however, MCI requires time linear only in the size of Delta, which is typically constant with respect to the size ....

G. Ramalingam. Bounded Incremental Computation. PhD thesis, Computer Sciences Dept., University of Wisconsin-Madison, 1993.

No context found.

Ramalingam, G., "Bounded Incremental Computation," Ph.D. dissertation and Tech. Rep. TR-1172, Computer Sciences Department, University of Wisconsin, Madison, WI (August 1993).

No context found.

Ramalingam, G., "Bounded Incremental Computation," Ph.D. dissertation and Tech. Rep. TR-1172, Computer Sciences Department, University of Wisconsin, Madison, WI (August 1993).

No context found.

Ramalingam, G., "Bounded Incremental Computation," Ph.D. dissertation and Tech. Rep. TR-1172, Computer Sciences Department, University of Wisconsin, Madison, WI (August 1993).

No context found.

G. Ramalingam. Bounded Incremental Computation, volume 1089 of Lec. Notes in Comp. Sci. Springer-Verlag, 1996.

No context found.

G. Ramalingam, Bounded Incremental Computation, Vol. 1089 of Lec. Notes in Comp. Sci., Springer-Verlag, 1996.

No context found.

G. Ramalingam. Bounded Incremental Computation, volume 1089 of Lec. Notes in Comp. Sci. Springer-Verlag, 1996. 18

No context found.

G. Ramalingam. Bounded Incremental Computation. PhD thesis, University of Wisconsin-Madison, 1993. 14

No context found.

G. Ramalingam. Bounded Incremental Computation. PhD thesis, University of Wisconsin-Madison, 1993.

No context found.

G. Ramalingam. Bounded Incremental Computation. PhD thesis, University of Wisconsin-Madison, 1993.

No context found.

G. Ramalingam. Bounded Incremental Computation. PhD thesis, University of Wisconsin-Madison, 1993. 14

No context found.

G. Ramalingam. Bounded incremental computation. In Lecture Notes in Computer Science 1089, 1996.

No context found.

G. Ramalingam. Bounded Incremental Computation. PhD thesis, University of Wisconsin-Madison, 1993.

No context found.

G. Ramalingam. Bounded Incremental Computation. PhD thesis, University of Wisconsin-Madison, 1993.

No context found.

G. Ramalingam. Bounded Incremental Computation. PhD thesis, University of WisconsinMadison, 1993.

No context found.

G. Ramalingam. Bounded incremental computation. Lecture Notes in Computer Science 1089, 1996.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC