### Table 5 Behavior Q

"... In PAGE 7: ... From the resulting state-action-behavior trajectories for the relearned behavior Q 6 , it is seen that most of the adaptation was con#0Cned to the lower level behaviors. This trajectory is shown in Figure 7 and Table5 . Fig- ure 3#28f#29 shows the #0Cnal hierarchical structure with the lower levels adapted to the new barrier.... ..."

### Table 1: Mobility Behavior

"... In PAGE 4: ... Hofbauer (1983) and Hofbauer and Nagel (1987), for instance, use data drawn from the Data Base of Employed Persons (which is used to keep track of worker apos;s social security earnings records) to follow all persons who successfully completed an apprenticeship in the middle of 1975 [ quot;75 cohort quot;], in the middle of 1979 [ quot;79 cohort quot;] and in the middle of 1984 [ quot;84 cohort quot;]. Table1 shows the fraction of each cohort that has left the apprenticeship rm, the apprenticeship sector and/or the apprenticeship occupation at the observation date (given in the row labeled quot;date quot;). The gures for sectoral and occupational mobility are derived from code-based measures of mobility.... In PAGE 26: ...Table1 0: Reduced-form multinomial logit estimates Dep.Var.... In PAGE 28: ...Table1 1: OLS estimates of wage equations Dep.Var.... In PAGE 29: ... Finally, Table 12 gives the results from estimating the structural multinomial logit model (2). Table1 2: Structural-form multinomial logit estimates Dep.Var.... In PAGE 31: ...3 are calculated in the usual way (i.e. estimating marginal e ects for each individual and then averaging over all individuals). Table1 3: Estimated marginal e ects of earnings Stayers (skilled job) Movers (skilled job) Movers (unskilled job) YSS .0012 -.... ..."

### Table 2. Agent Behavior, With Adaptation

2004

"... In PAGE 10: ...2 Adaptive Tree Mechanism We then made use of the adaptive tree mechanism to select the best available machines for the torus in a decentralized manner. The behavior of each agent was as in Table2 . The feedback sent by each child to its parent was the time taken by the child to complete its two previous tile multiplications.... ..."

Cited by 10

### Table 7 Agent behavior, with adaptation.

2004

"... In PAGE 12: ...2 Adaptive Tree Mechanism We then made use of the adaptive tree mechanism to select the best available machines for the torus in a decentralized manner. The behavior of each agent was as in Table7 . The feedback sent by each child to its parent was the time taken by the child to complete its two previ- ous tile multiplications.... ..."

Cited by 10

### Table 2. Behavior of the expansion of the prime 61 relative to di erent bases, compared with prime 31 ergodic to base 3.

"... In PAGE 11: ... Each line is a period, but it is not a minimal period except in the ergodic cases b = 2; 6; 7; 10 and 30. In the ergodic cases b = 2; 6; 10; 30 we see from Table2 that each single digit 0; 1; : : : ; b?1 occurs exactly the same number of times, that is, we have exact equidistribution of single digits. This follows from the equidistribution theorem and the (accidental) fact that bjp ? 1 for the cases chosen.... In PAGE 13: ...1=61 in Table2 , in fact, it is the binary expansion of 234=61 (mod 1) = 29=61. Irrespective of whether 61 is ergodic or not, all sequences in Table 2 ex- hibit certain intuitively acceptable features of randomness; of course, they also conform with our de nition of randomness.... In PAGE 13: ...1=61 in Table 2, in fact, it is the binary expansion of 234=61 (mod 1) = 29=61. Irrespective of whether 61 is ergodic or not, all sequences in Table2 ex- hibit certain intuitively acceptable features of randomness; of course, they also conform with our de nition of randomness. This new class of random sequences of digits has one immediate practical application: it enables precise statements to be made in a debate with the \fanatical school quot; of probability theory.... ..."

### Table 1: Some basic functions, their domain, range, behavior and inverse.

2005

"... In PAGE 4: ... When f(x) = f( x), for all x 2 R, then f is called odd. Table1 shows the basic functions usually studied in math at high school, if we exclude the trigonometric functions (sine, cosine, tangent, co-tangent), generic polynomial functions polan;:::;a0 : x 7! Pn i=0 aixi and the quadratic function pola;b;c : x 7! ax2 + bx + c. The composition, sum, di erence, product and quotient of functions (herein represented by , +, , , =) are also studied, which enable the use and construction of more complex functions.... In PAGE 4: ... We note also that although pk = ck id, we are going to keep all three as basic functions, an usual practice in high school. Table1 is seen as useful domain knowledge that students learn and thus the solver shall also refer to. Working as a knowledge base, it may be extended if appropriate and also reduced.... In PAGE 9: ... 4.1 Computing the domains of the involved expressions In the implementation of the predicate that computes the domain of a function, the infor- mation given in Table1 is translated to clauses and consulted when required. Then, the domain is computed by actually solving some problems, so that the solving procedure and domain determination are not independent.... In PAGE 13: ... By the same rule, it deduces that h(abs rad3 pol 1;2)(x) 0; x; Ri can be rewritten in solved form hx 2 R; x; Ri. For a basic implementation of the solver, we may con ne the application of this rule to the cases when the required information is explicitly given in Table1 . However, some math problems are often solved just with a glance, and, clearly, a brief reasoning.... In PAGE 16: ... This is the case of abs and pow2n, for n 1. It is interesting also to see that, from Table1 and Rules 6 and 7, we may automatically infer that, for instance, h(pola;b f)(x) P k; x; Di hf(x) P (k b)=a; x; Di if a gt; 0. h(pola;b f)(x) P k; x; Di hf(x) P 1 (k b)=a; x; Di if a lt; 0.... ..."

### Table 3: Behavior under adaptive granularity

2005

"... In PAGE 13: ... The developers were quite interested in the race reports. We added performance counters to investigate the inter- nal actions of several of the test programs under adaptive granularity, with the results shown in Table3 . (Including these counters degrades performance significantly, so they are not present in the data in Table 2.... ..."

Cited by 28

### Table 3: Behavior under adaptive granularity

2005

"... In PAGE 13: ... The developers were quite interested in the race reports. We added performance counters to investigate the inter- nal actions of several of the test programs under adaptive granularity, with the results shown in Table3 . (Including these counters degrades performance significantly, so they are not present in the data in Table 2.... ..."

Cited by 28

### Table 6 Agent behavior, without adaptation.

2004

Cited by 10

### Table 2 Behavior Q

"... In PAGE 6: ... The two existing behaviors quickly integrated the new behav- iors into their action sequences. The hierarchical struc- ture of the control system is further shown in the new action sequence when behavior behavior Q 6 was acti- vated is shown in Table2 and Figure 4. This shows that the sub-behaviors, Q 7 and Q 8 are evident in the state- action-behavior trajectories of the higher level behaviors Q 5 and Q 6 #28not shown#29.... ..."