Semantics of Instruction Heads and Their Arguments

In what follows, $x_1,x_2,$ $y_1,y_2,$ $z_1,z_2,$ $\in \{ 0, \ldots, n-1 \}$ stand for instruction arguments selected according to probability distributions $Q(IP + 1, .), Q(IP + 2, .), \ldots, Q(IP + 6, .)$, respectively. They are used to address state components and module columns. For simplicity, instruction descriptions below use the following macros: $x := (x_1 n + x_2)~mod~m;$ $y := (y_1 n + y_2)~mod~m;$ $z := (z_1 n + z_2)~mod~m.$

Back to Active Learning - Exploration - Curiosity page