### Table 4: Grid of main classes which underpinned the final selection

2007

"... In PAGE 8: ... To counter that, the 48 clips were deliberately selected to cover material showing emotion in action and interaction; in different contexts (static, dynamic, indoor, outdoor, monologue and dialogue); spanning a broad emotional space (positive and negative, active and passive) and all the major major types of combination (consistent emotion, co-existent emotion, emotional transition over time); with a range of intensities; showing cues from gesture, face, voice, movement, action, and words and representing different genders and cultures. Table4 shows the grid of main classes which underpinned the final selection. Each clip was chosen for its representation in those main classes.... ..."

### Table 2: Timing and causality.

1998

"... In PAGE 8: ... We propose here a framework which generalizes these notions (we follow the technique presented in [4]). In Table2 , we associate to each Hybrid primitive statement, its presence calculus and causality calculus. The pres- ence calculus involves only presences and boolean signals, it summarizes the constraints on presences and their relations to boolean signals.... In PAGE 8: .... The statement \on T quot; is optional. When it is omitted, then causality X gt; Y always hold. Table2 is commented now. For a signal X, the status of X a instant t shall be \absent quot; if t 62 TX, and the value of Xt otherwise.... In PAGE 10: ...= TY = T (1) k S = T n U (2) k U = true( pre(Y ) init 0 0 ) (3) k T gt; (X; Y; pre(Y )) (4) k U gt; U (5) k (T ; U) gt; S (6) k on T : pre(Y ) gt; U , T gt; U (7) k on U : U gt; Y , U gt; Y (8) k on S : (X; Y ) gt; dY=dX , S gt; dY=dX (9) Table 3: The example of table 1 : presence and causality calculus. This Hybrid program is obtained by applying the rules of Table2 until xpoint is reached. To apply these rules, we should have introduced the intermediate signal Z = f(X; Y ) ; this intermediate signal is simply denoted by dY=dX in the above program.... In PAGE 10: ... Selfexplanatory. Table2 is used as follows for executing a program. First, for each statement of the program, add to this program the associated presence and causality calculi, following Table 2.... In PAGE 11: ... This cannot be avoided in general, but is better performed most seldomly. Since causality constraints, as derived from Table2 , depend on the syn- tax of the program, we can try to rewrite this program di erently, with the objective of breaking most possible circuits [4]. Rewriting part of the pro- gram involving timing and booleans, can be performed using exactly the techniques of Signal language compilation, since the Signal clock calcu- lus has identical algebraic structure as our presences and boolean signals.... In PAGE 14: ...ignal at instant t, i.e., some preorder on the set fX1,: : : ,Xkg. 3. For X; Y signals of the considered hybrid system P , we de ne on T : X gt; Y (15) as X t Y holds 8t 2 T (16) which consequently requires T TX \ TY : (17) The rationale for infering causality constraints in Table2 is simple. Actions of computing are abstracted as term rewriting : the computation action y = f(x) is abstracted as \y can be substituted by f(x) quot;, which is encoded as the preorder x y, also written x gt; y.... ..."

Cited by 10

### Table VII Granger Causality

2001

Cited by 35

### TABLE 1 Causal Heuristics

### Table 3: Complexity of Probabilistic Causality

"... In PAGE 20: ... We also analyze the complexity of computing the probability of a causal formula. Our complexity results are summarized in Table3 . In detail, deciding probabilistic causal irrelevance is Ca30... ..."

### Table 2: Complexity of Event Causality

2001

"... In PAGE 4: ... 4.2 Results Our complexity results for the above causal relationships be- tween events are summarized in Table2 . We distinguish be- tween the general and the binary case, in which we assume a syntactic restriction to binary causal models.... ..."

Cited by 15