R. Rubinfeld. Designing checkers for programs that run in parallel. Algorithmica, 15(4):287-301, 1996.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....itself to see if it is identical to the original one. To speed up the test, a message digest algorithm such as MD5 [71] can be used. 2. We can examine the validity of intermediate results produced by the program. This technique is known as program (or result) checking [12] 13] 30] 73] [74], 95] and has been touted as an alternative to program verification and testing. 3. We can encrypt the executable, thereby preventing anyone from modifying it successfully unless they are able to decrypt it. The decryption routines must be protected from reverse engineering by hardware means, by ....

R. Rubinfeld, "Designing Checkers for Programs that Run in Parallel," Algorithmica, vol. 15, no. 4, pp. 287-301, Apr. 1996.

....the executable program itself to see if it is identical to the original one. To speed up the test, a message digest algorithm such as MD5 [71] can be used. 2. We can examine the validity of intermediate results produced by the program. This technique is known as program (or result) checking [12,13,30,73,74,95] and has been touted as an alternative to program verification and testing. 3. We can encrypt the executable, thereby preventing anyone from modifying it successfully unless they are able to decrypt it. The decryption routines must be protected from reverse engineering by hardware means, by ....

Ronitt Rubinfeld. Designing checkers for programs that run in parallel. Algorithmica, 15(4):287-- 301, April 1996.

....can extend AgoVista with problems with which they are familiar, or with references to new algorithms they have developed for these problems. AgoVista is based on a novel application of a technique known as program (or result) checking, developed over the last decade by Manuel Blum and others [4 6,10,16,17,19] as an alternative to program verification and testing. Program checking extends programs with checkers to allow them to verify the correctness of the results they compute. 1.1 Two Motivating Episodes To motivate the need for specialized search engines for computer scientists, we will consider ....

....possibly return a list of compilation errors or security violations that have to be fixed before the checklet will be accepted into the coop. The most difficult part is certainly constructing the actual program checking code. Inspiration can sometimes be had from the result checking literature [4 6,10, 16,17,19], but more often by examining existing checklets. AgoVista supports this by making the source code of checklets available for perusal. 11 Figure 8: Query transformations (A) represents Null. Greek letters are type variables. Examples are in the format signature:query. T A ....

Ronitt Rubinfeld. Designing checkers for programs that run in parallel. ALGORITHMICA, 15(4):287-- 301, April 1996.

....is its self testing correcting pair. Related work. To the best of our knowledge, this is the first time that the selftesting correcting of protocols is discussed. Program checking and self testing have been extensively investigated, and checkers and self testers are now known for many problems [10, 2, 19, 26, 27, 17]. Other relevant works include checking the correctness of memory [6] cryptographic program checking [18] and batch checkers [8] In the related but different area of self stabilizing protocols (e.g. 1, 3, 15, 22] the notion of checking has been considered as an important co routine. This ....

R. Rubinfeld. "Designing Checkers for Programs that Run in Parallel." Algorithmica 15(4), pp. 287-301, 1996.

....the executable program itself to see if it is identical to the original one. To speed up the test, a message digest algorithm such as MD5 [40] can be used. 2. We can examine the validity of intermediate results produced by the program. This technique is known as program (or result) checking [7 9, 18, 41, 42, 52] and has been touted as an alternative to program verification and testing. 11 3. We can generate the executable on the fly, in the hope that even minor changes to the generating program will produce code that cannot be executed. Aucsmith [4, 5] was the first to suggest 3. Intel s ....

Ronitt Rubinfeld. Designing checkers for programs that run in parallel. ALGORITHMICA, 15(4):287--301, April 1996.

....Result Checking Against Adaptive Programs and in Cryptographic Settings (Extended Abstract) Manuel Blum Computer Science Division U.C. Berkeley Berkeley, California 94720 Michael Luby International Computer Science Institute Berkeley, California 94704 Ronitt Rubinfeld y Computer Science Division U.C. Berkeley Berkeley, California 94720 May 17, 1990 Abstract The theory of program result checking introduced in [Blum] allows one to check that a program P correctly computes the function f on input x. The checker may use P s outputs on other inputs to ....

....have been found are also adaptive checkers. For example, it can easily be seen that the GCD checker in [Adleman Huang Kompella] and that all of the checkers given in [Blum Kannan] are adaptive. Other checkers do not work for an adaptive program. Examples of such checkers are the ones in [Blum Luby Rubinfeld] where adaptive programs can easily fool the checkers. At the present time we see no way to convert such a checker into an adaptive checker. However, if more than one copy of the program exists, we show that a checkers based on the methods in [Blum Luby Rubinfeld] can work for adaptive ....

[Article contains additional citation context not shown here]

Rubinfeld, R. "Designing checkers for programs that run in parallel", Manuscript.

....instance to several random instances and computing the answer to the given instance from the answers to the random instances. Lipton[29] building on the work of Beaver and Feigenbaum[4] shows how polynomials in general and the permanent in particular are amenable to this technique. Rubinfeld[34] extends the notion of checking to parallel checking while Blum et al. 8] extend it to programs that store and retrieve data from unreliable memory. Kannan and Yao[24] have considered the problem of checking coin tossing programs that produce specified output probability distributions. There are ....

R. Rubinfeld. Designing Checkers for Programs that Run in Parallel. Tech. Report TR-090-040, International Computer Science Institute, Berkeley, August, 1990.

....program not just on x but also on other points. If P is correct, meaning it correctly computes f at all points, then C must accept x, but if P (x) 6= f(x) then C must reject x with high probability. Program checking has been extensively investigated, and checkers are now known for many problems [7, 1, 6, 16, 8, 21, 22, 13]. Checking has also proven very useful in the design of probabilistic proofs [23, 2] Batch program checking was introduced by Rubinfeld [20] Here the checker gets many instances x 1 ; xn . Again if P is entirely correct the checker must accept. And if P (x i ) 6= f(x i ) for some i the ....

R. Rubinfeld. Designing Checkers for Programs that Run in Parallel. Algorithmica, 15(4):287--301, 1996.

No context found.

Rubinfeld, R., "Designing Checkers for Programs that Run in Parallel", ICSI technical report TR-90-040.

No context found.

R. Rubinfeld. Designing checkers for programs that run in parallel. Algorithmica, 15(4):287-301, 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