Results

**1 - 5**of**5**### Table 1: The Dena apos;ina verb template (adapted from Tenenbaum 1978)5

"... In PAGE 3: ... The iterative morpheme nu occurs in position 16 of Tenenbaum apos;s (1978) model of the Dena apos;ina verb template (cf. Table1 ).4 This position is in the disjunct complex, to the right of the ... ..."

### Table 1: Simulation parameters Spatial Frequencies fi corresponding Standard deviations

"... In PAGE 9: ... Cell Parameters We separately simulate experiments performed in central cat area 17 (0-5 degrees eccentric- ity) and more peripheral area 17 (8-12 degrees eccentricity). In each case, we use spatial frequency and position shift distributions derived from published data for these regions, see Table1 . We also simulate results from some experiments where RFs were mapped by reverse correlation (DeAngelis et al.... In PAGE 17: ... Fig. 6A shows the distributions of peak disparities predicted by three models for a set of tuned binocular cells simulated with parameters chosen to be representative of ocularly- balanced cells in the central visual eld of cat area 17 (see Table1 , Methods). In the subregion correspondence model, this distribution is bimodal, with a narrow peak near D = 0, a near absence of cells tuned to 0.... In PAGE 18: ...Table1 | the distribution of tuning curve peaks is broader than the experimental data, Fig. 6Ab (only 52 % of cells have preferred disparity lt;= 0:25 ).... In PAGE 20: ...ion. This follows from the fact that from Eq. 3, xi is measured parallel to the V -axis for horizontal-preferring cells ( i = =2) but parallel to the H-axis for vertical-preferring cells ( i =0). Joshua and Bishop (1970) reported such an anisotropic distribution of position shifts in cat area 17, with a wider distribution of horizontal than vertical position shifts, at eccen- tricities of 8{12 degrees near the horizontal meridian; whereas in the central 4 degrees of the visual eld, they reported an isotropic distribution of position shifts (see Table1 and Methods). In Fig.... In PAGE 20: ... For comparison, we show in Fig. 7c{d the results of simulations using data t to independently measured distributions of position shifts and spatial frequencies from central (0 {4 ) and more peripheral (8 {12 ) parts of area 17 (see Table1 , Methods). Based on these measured distributions, the alternative explanation allowed by subregion cor- respondence predicts that phase shifts should be evenly distributed as a function of preferred orientation for central locations, Fig.... In PAGE 40: ... 8. Spatial frequencies and position shifts were chosen to correspond to central area 17, see Table1 , Methods, except that the position shifts in (b) were all set to zero. Since the distributions are symmetric with respect to positive and negative disparities, only positive values are shown.... In PAGE 40: ... Since the true position shifts in that data are unknown, position shifts are drawn from the anisotropic distribution measured in \Mildly Peripheral (8 {12 ) quot; area 17. Use of the more peripheral, rather than central, position shift distribution is indicated by the low spatial frequencies observed in the experimental data, see Table1... In PAGE 41: ...f data from 100 cells in which subregion correspondence was imposed by Eq. 8. As in Fig. 7b, spatial frequencies were modeled on those observed in the experimental \Reverse Correlation quot; distribution, while position shifts were modeled on those observed in \Mildly Peripheral (8 {12 ) quot; area 17, see Table1 . The simulations in (d){(e) were based on a di erent set of data simulated with identical parameters to (a){(c) except that left- and right-eye phases were independently chosen.... ..."

### Table 4.1 Measured Average Search Costs of Several Algorithms The apos;P apos; and apos;T apos; in the le names indicate whether the le is Pascal source or English text. OSO stands for the expected cost if the records are left in the optimal static ordering, and MTF is the move-to-front algorithm. The mark for Zipf indicates the expected cost of an optimal static ordering, assuming that the probabilities follow Zipf apos;s distribution. Bentley and McGeoch compare this to the expected cost of the optimal static ordering to suggest that the probabilities of the words are closer to the Zipf distribution than a normal distribution (where the expected cost would be about half the number of distinct words in the le). The following table submitted by Tenenbaum [1978] contains a condensation of a more extensive table of simulations on move-ahead-k algorithms. Files of various sizes (N) containing data which was distributed by Zipf apos;s law were accessed 12; 000 times, for 1 k 7. The best value of k (among those tested) is given for the sizes tested.

1985

Cited by 20

### Table 2. Simulated realistic processor parameters (per processor core) processor parameter value

"... In PAGE 14: ... The spawn queues, endblock queues, memory sharing mechanism, and register sharing mechanism, as described in Section 4, are implemented as modules in Joshua. Table2 gives the parameters for the baseline uniprocessor. Table 3 gives additional parameters for the corresponding control/data parallel CMP.... ..."

### Table 1. The 8 and 9 cluster number models.

2002

"... In PAGE 11: ... On the basis of figure 2, two of the more impressive models appear to be those which exhibit the best data-fit using 8 and 9 clusters. These models are detailed in Table1 , which lists the clusters, their associated weights, the additive constant, and the percentage of variance in the data explained. This information confirms that, as suspected on the basis of their equivalent data-fit, the 8 cluster models generated by the new method and by Tenenbaum (1996) are identical.... ..."

Cited by 10

**1 - 5**of

**5**