Richard Martin. Personal communication, August 1995.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....the azimuthal perturbation. The phase shift fi is set to zero. The amplitude ej 0.2 given by [Martin 85 Meiburg (1992) based on linear stability arguments corresponds to a perturbation of 4 . For the optimization, we implement a (1 i) ES running for 300 generations with initial values given by [Martin (2001)] With the phases constrained between 0 and 1, two optimizations are performed. The first optimization, for which the amplitudes are constrained between 0 and 1, yields an increase of 42 for the filament length. In the second optimization with lower and upper amplitudes of 0 and 0.25, ....

J.E. Martin, Personal communication, 2001.

....gen :FWk:R :I of Exercise 2.9.15b. e) Show that an funB]R: on a recursive datatype can bewritten as a paramorphism: h = para (h # in # Fexr) Thus, paramorphisms are extremely gen R]I 16.On the codatatype of lists from 2.6. 7,defin asan un # the funR 5] inT , such that inT (1, 5) [1, 2, 3, 4, 5] in (5, 5) 5] inT (6, 5) 17.On the codatatype Stream A = coda ta (A) the fun#B I iterate isdefinI by iterate f =unfol Stream (id # f ) Usin unn fusion prove that map f # iterate f = iterate f # f 18. For codatatype T = coda ta F, Uustaluan Ven [40, 38] dualize ....

.... pairs.Then the state of theiteration fordiagon conbnB of a pair, afinR] list of those streamsseen so far an a stream of streamsnr yetseen At each step, strip an :R: diagonW o# from the streamsseen so far,an inBI:F anI:F stream from thosen ot yet seenF This example is due to Richard Bird [3]. 4 Applic61 We con:#R] these lecturen otes with three more substan tial examples of the con 5:F we have described: a simple compiler for arithmeticexpressionR laws formon]R an comon:Wk an e#cien t programs for breadth first traversal of trees. 184 Jeremy Gibbons 4.1 A simple compiler In ....

Richard Bird. Personal communication, 1999.

....e and d are uniquely determined by universality (as they turn out to be the morphism into the terminal object and diagonal in the co Eilenberg Moore category) regardless of the adjunction used to identify them. Another concise version of the de nition of a linear category has been proposed in [14]. There the property chosen as de ning a linear category is the fact that the co Eilenberg Moore category is a category of comonoids. Like ours, this de nition is completely equivalent to the original one. The next proposition explains our choice of morphisms in the previous section, which did ....

....this de nition is completely equivalent to the original one. The next proposition explains our choice of morphisms in the previous section, which did not mention any preservation for e and d. It also shows that our de nition of morphism provides a linearly distributive functor in the sense of [14]. Proposition 21. If (K; m ; m I ) S S 0 is a Lin morphism, then Ke = m I e 0 K and Kd = m d 0 K. Proof. This is essentially based on proposition 20 and on the fact that K = 0 K as monoidal functors. 5.2 DILL Models rede ned In this section we rede ne the ....

Martin Hofmann and Thomas Streicher. Personal communication, 1999.

....be 1 if x satis es the ith term of g and 0 if x does not. At most log(r) variables are relevant for P i and thus the Fourier representation of P i has no more than r non zero coecients. Using the principle of 2 A similar result has also been shown by Lipton using a somewhat di erent analysis [16]. 6 inclusion exclusion we can create a function P 0 from the P i s which is (rs 0 ) k sparse and which is 1 when g is satis ed and 0 otherwise. Speci cally, let P 0 = X i 1 P i 1 X i 1 i 2 P i 1 P i 2 X i 1 i 2 i 3 P i 1 P i 2 P i 3 ( 1) k X i 1 i k P i 1 ....

Richard Lipton. Personal communication.

....to order the nodes hence any random order is as good as any other. 3 To get the average run time, each 3 As an aside, the reader should note that the ordering information does not always improve the effectiveness of DFS. For example, experience with the IDA algorithm on the 15 puzzle problem [11] indicates that 7 experiment is repeated many times. Note that, besides the random ordering of successors, there is another source of variability in execution time of parallel DFS. This is because the parts of the state space tree searched by different processors are determined dynamically, and ....

Richard Korf. Personal Communication. 1988.

....to form a symmetric monoidal adjunction. We write for the left adjoint. Note that I is a monoidal functor by using the internal language. It is also possible to define the exponentials by the condition Hom E( Gamma ) A; B) Hom E( Gamma DeltaA) I ; B) plus a Beck Chevalley condition [HS99]. It is easy to see that these two definitions are equivalent: if you specialise the second condition to the case Gamma = and use the adjunction between B and Gr(E) putting ( A) j (A) you obtain the first condition by the first requirement of restricted indexed categories. The converse ....

Martin Hofmann and Thomas Streicher. Personal communication, 1999.

.... thought that all useful designs have exactly one tSCC [61] on the other hand, other people have argued that if we pick an arbitrary piece of logic from a larger design, this piece may have more than one tSCC even though the larger design has only one tSCC representing the steady state behavior [66]. In this thesis we are motivated by the problem of replacing arbitrary pieces of logic and thus we allow designs to have multiple tSCCs. CHAPTER 5. SYNTHESIS AND VERIFICATION FOR SAFE REPLACEABILITY 48 procedure tSCC (input: T ( x; i; y) f A( x) onion ring(T ( x; i; y) 1) forever ....

....deal with the validity of retiming transformations with the assumption that all latches are no reset latches. For designs where some latches are reset latches we can transform these latches to no reset latches using the transformation described in Chapter 2. However, it has been pointed out to us [66] that once we transform an original design to a new design via such transformations we may no longer be able to retime the new design to designs which were retimed versions of the original design. In [54, page54] Malik also claims that making the reset circuitry explicit, as we suggest by our ....

Richard L. Rudell. Synopsys, Inc., Mountain View, CA. Personal communication, 1995.

....Labs recently compared the speed of three stack distance programs: The first author s publicly available implementation of the fast splay tree based priority depth algorithm of Section 3. 1 [25] a simple O(MN) linked list implementation supplied with Kelly s fast code, and Arlitt s own program [3]. Arlitt s implementation divides the LRU stack into a number of equal sized bins, each of which contains 50 items. The advantage of this approach is that the worst case number of operations to process a reference is proportional to the number of bins plus the number of items in a bin, rather ....

Martin Arlitt. Personal communication.

....the cluster. 5. Replacing mesh subdivision by textures [19, 10] is an effective means of avoiding the cost due to the subdivided hierarchy itself. In the case of time dependent illumination, texture movies could be generated for this purpose, although a specific algorithm has yet to be proposed [13]. 6. The use of a shooting approach [20] would eliminate the need to store all the links in the solution, especially the very many links from primary light sources. 7. It would be interesting to compare this algorithm to Drettakis dynamic update technique based on the line space hierarchy, but ....

Ignacio Martin. Personal Communication.

....the cluster. 5. Replacing mesh subdivision by textures [19, 10] is an effective means of avoiding the cost due to the subdivided hierarchy itself. In the case of time dependent illumination, texture movies could be generated for this purpose, although a specific algorithm has yet to be proposed [13]. 6. The use of a shooting approach [20, 10] would eliminate the need to store all the links in the solution, especially the links from primary light sources. 7. A precise comparison of this algorithm with Drettakis and Sillion [5] dynamic update technique based on the line space hierarchy ....

Ignacio Martin. Personal Communication.

....additional reference microphone positioned close to the talker is often provided to obtain high quality Z. speech for transcription e.g. Pallett et al. 1995 . In addition, a committee of motivated listeners is often used to transcribe some of the more difficult materi Z. als e.g. Leonard, 1984; Martin, 1996 . If it is assumed that the transcription word error rate is lower than reported word error rates for intelligibility testing with human listeners, then the transcription error Z rate is less than 0.009 for read digits Leonard, 1984 , less than 0.4 for read sentences from the Z. Wall Street ....

....testing with human listeners, then the transcription error Z rate is less than 0.009 for read digits Leonard, 1984 , less than 0.4 for read sentences from the Z. Wall Street Journal Ebel and Picone, 1995 , and less than 4 for spontaneous conversations recorded Z. over the telephone Martin, 1996 . 4. Performance comparisons Sections 4.1, 4.2, 4.3, 4.4 and 4.5 summarize recent comparisons between the speech recognition performance of humans and machines using the six speech corpora shown in Fig. 1 supplemented by other speech materials that have been used to evaluate human performance. ....

[Article contains additional citation context not shown here]

A. Martin, 1996. Personal communication.

....#Euler Time [JTS89] 10 22 19 20 0.00 1 5 7 1 0 [K92] 45 85 80 82 0.00 24 56 83 1 23 [C92] 60 165 164 164 0.00 0 2 27 1 4 [H93b] 20 30 28 0.00 1 8 16 1 1 [H93a] 34 45 43 0.00 1 4 5 1 1 [TBB88] 43 62 58 0.00 46 213 257 1 60 [H93a] 46 64 62 0.00 1 5 14 1 1 [H93a] 48 69 64 0. 00 258 1009 705 1 327 [M93] 17 39 35 0.00 1 2 0 11 0 [M93] 30 56 53 0.00 8 23 28 1 3 [M93] 45 98 88 2.20 1090 3312 416 144 1000 [M93] 47 99 91 0.00 1234 3365 406 1 984 [M93] 47 101 89 4.30 564 2490 719 130 1000 [M93] 61 187 130 22.62 206 1147 441 1873 1000 [FR76] 8 24 113 113 0.00 1 2 0 3 0 [F92] 8 28 1982 1982 0.00 8 30 15 ....

....19 20 0.00 1 5 7 1 0 [K92] 45 85 80 82 0.00 24 56 83 1 23 [C92] 60 165 164 164 0.00 0 2 27 1 4 [H93b] 20 30 28 0.00 1 8 16 1 1 [H93a] 34 45 43 0.00 1 4 5 1 1 [TBB88] 43 62 58 0.00 46 213 257 1 60 [H93a] 46 64 62 0.00 1 5 14 1 1 [H93a] 48 69 64 0.00 258 1009 705 1 327 [M93] 17 39 35 0. 00 1 2 0 11 0 [M93] 30 56 53 0.00 8 23 28 1 3 [M93] 45 98 88 2.20 1090 3312 416 144 1000 [M93] 47 99 91 0.00 1234 3365 406 1 984 [M93] 47 101 89 4.30 564 2490 719 130 1000 [M93] 61 187 130 22.62 206 1147 441 1873 1000 [FR76] 8 24 113 113 0.00 1 2 0 3 0 [F92] 8 28 1982 1982 0.00 8 30 15 12 1 [L92] 10 44 1101 1105 ....

[Article contains additional citation context not shown here]

Martin, A., "Personal communication", (1993)

....invalid) so future modifications to that page on other nodes will not send additional updates to the node. This reduces network tra#c for pages which are no longer used by a given node. The threshold used to determine when to expire a page in this situation is determined by the following formula [17]: C T u = T p (3.1) The value C is the count of the number of updates sent to the node for the page in question, T u is the time required to send an update for one word, and T p is the time needed to copy an entire page from another node. Since the latter two are fixed by the architecture, ....

Richard P. LaRowe Jr. Personal Communication.

....on conservative and invertible primitives will have very low heat dissipation. Invertibility is related to the idea of a group, where each element of a group has an inverse. Logic synthesis based on conservative and invertible functions and group symmetries have been recently investigated by Morf [55]. New results from this type of work may lead to the design of more general instructions and datapaths for many symmetry related operations. Further improvements in propagation mechanisms are likely to focus on the memory bandwidth issue. Currently, ALGE s cycle time is limited by the page mode ....

Martin Morf, 1993. Personal communication.

No context found.

Richard Martin. Personal communication, August 1995.

No context found.

Martin P. Robillard, May 2005. Personal communication.

No context found.

J. Martin, "Personal Communications," Griner and Schmitz, Kansas City, MO, June 2002.

No context found.

Richard Beigel. Personal communication. 2003.

No context found.

Martin Hirt. personal communication. Jul 1999.

No context found.

Richard Lipton. Personal communication.

No context found.

Richard Beigel. Personal communication. 2003.

No context found.

J.E. Martin, Personal communication, (2001). 5

No context found.

Goehlert, Richard. 1999. Personal Communication. U.S. EPA. Boston, MA.

No context found.

R. Martin, UC Berkeley. Personal communication, April 1999. 179

No context found.

Richard Chang, 1997. Personal communication.

First 50 documents

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