### Table 2. BKYY DR for ICA and Blind Source Separation (BSS)

1997

Cited by 19

### Table 1. The dimensions dr ; ^ dr ; dr ; lr for low r.

"... In PAGE 13: ... Further, the linear mapping f 7! f maps a minimal left ideal either onto 0 or onto an equi- valent minimal left ideal. In Table1 we have listed the rst values of the dimensions dr ; ^ dr ; dr ; lr. Since these dimensions are monotonically increasing functions of r and I(r) has a subideal of dimension dr for all r 1, we read from Table 1 that for r 4 subideals of dimensions ^ dr ; dr can not occur in I(r).... In PAGE 13: ... In Table 1 we have listed the rst values of the dimensions dr ; ^ dr ; dr ; lr. Since these dimensions are monotonically increasing functions of r and I(r) has a subideal of dimension dr for all r 1, we read from Table1 that for r 4 subideals of dimensions ^ dr ; dr can not occur in I(r). Moreover, for r = 3 a subideal of I(r) of dimension ^ d3 = 35 is impossible.... ..."

### Table 4: The Values of dr for PVA Models OBS dr for dr for

1997

Cited by 1

### Table 1: Particular values of c and q at the normal model.

1999

"... In PAGE 5: ... It is shown in Section 2 of BDJ that for this distribution the MCD-problem has a unique solution given by the ellipsoid A(F ; ) = fz 2 IRpj(z ? )t ?1(z ? ) q g (2.11) where G(t) = PF0;I(ZtZ t) and q = G?1(1 ? ), with corresponding MCD-functionals T (F ; ) = (F ; ) = c 1 ? Z ztz q z2 1 dF0;I(z) : To obtain Fisher-consistency at this model, it su ces to set c = 1 ? Rztz q z2 1 dF0;I(z) = (1 ? ) ( p2 ?(p 2 + 1) Z pq 0 rp+1g(r2) dr )?1 : Table1 gives several values of the constants c and q for di erent values of p and at the normal model. Since the MCD is a ne equivariant we will only derive the in uence function at the model distribution F = F0;Ip.... In PAGE 22: ...Table1 0: Median squared error and 0.9 quantile of the squared errors (between parenthesis) of element (1,2) of the MCD, S and reweighted MCD and S scatter matrix estimators at a 20% contaminated normal distribution.... ..."

Cited by 11

### Table 9. Values of dr and ds for various FACE*CONDITION combinations whole mouth eyes ds dr ds dr ds dr

in Recognising Emotions in Human and Synthetic Faces: The Role of the Upper and Lower Parts of the Face

"... In PAGE 5: ... Hence, dr signals the amount of uncertainty as to the identity of the stimulus originating the response. Table9 reports the values of ds and dr for the various FACES*CONDITIONS combinations. Table 9.... ..."

### Table 8: Reasons for Not Documenting DR

2005

"... In PAGE 13: ...Table 8: Reasons for Not Documenting DR Table8 summarizes the responses to the reasons for not documenting DR. These results reveal that lack of time/budget (60.... ..."

### Table 1: Parameters the input rate should equal the output rate, else every period we would accumulate more and more data in memory. Thus, we have the equation S = DR T: (1) In a feasible system, the period T must be large enough so that all necessary IOs can be performed. Since T is xed and cannot vary depending on the number of concurrent requests in the system at a particular moment, we must make T large enough to accommodate NLimit seeks and transfer NLimit segments. Furthermore, in computing these seek times, we have to assume a worst case situation, so that no matter where the segments are located on disk, we will have enough time to read them. The total seek overhead for NLimit requests is PNLimit i=1

"... In PAGE 3: ... Sweep implements an admission control policy that turns away requests when the system is already handling NLimit requests. To assist the reader, Table1 summarizes these parameters, together with other parameters that will be introduced later. The rst portion of Table 1 lists the basic xed and tunable pa- rameters.... In PAGE 3: ... To assist the reader, Table 1 summarizes these parameters, together with other parameters that will be introduced later. The rst portion of Table1 lists the basic xed and tunable pa- rameters. The second portion describes subscripted parameters that are used for characteristics of individual requests.... ..."