RMIT Discrete Mathematics Functions Guide
RMIT Discrete Mathematics Functions Guide
A set of ordered pairs is a function from set X to set Y if every element in X is assigned exactly one element in Y. This can be established using arrow diagrams, where each element in X is connected by an arrow to one and only one element in Y. In situation (a) from Source 1, the set {(1, a), (2, a), (3, c), (4, b)} is considered a function because each element in X = {1, 2, 3, 4} has a unique arrow pointing to an element in Y = {a, b, c, d}. This can be visualized through an arrow diagram that represents these mappings. In contrast, in situation (b) where the set is {(1, c), (2, a), (3, b), (4, c), (2, d)}, there are two outputs for the input 2 (namely a and d), which violates the definition of a function .
Composition of two functions f and g can lead to different properties regarding injectivity and surjectivity based on the individual characteristics of f and g. For instance, if both f and g are onto functions, then the composition f ◦ g is also onto as each element in the codomain of f ◦ g can be sourced from an element in the domain of g due to both functions covering their codomains. However, if f ◦ g is one-to-one, this does not imply that f is one-to-one, as shown in the counterexample where g : {1} → {a, b} given by g(1) = a and f : {a, b} → {α} given by f(a) = f(b) = α, resulting in f ◦ g as one-to-one but f not being one-to-one .
In telecommunications, the number of ATM cells transmitted over a specified time is determined by dividing the total data that can be transmitted by the size of each cell. For a connection transmitting at 500 kilobits per second and an ATM cell size of 53 bytes (i.e., 424 bits), in a timeframe of 60 seconds, the total bits transmitted would be 30,000,000. The number of cells transmitable in this period is the integer division of total bits by the size of one cell, calculated as 30,000,000/424, resulting in 70,754 cells transmitted per minute, considering the remainder is disregarded as partial cells cannot be transmitted .
In matrix-vector multiplication-based functions like f : R^3 → R^3 defined by f(a) = Aa, injectivity is determined by checking if the null space of the matrix A contains only the zero vector, meaning f(x) = f(y) implies x = y, thus confirming injectivity if this is met. Surjectivity is verified by solving f(x) = a for any vector a in R^3 to check if solutions always exist, indicating that every possible output is accounted for by some input. In the example provided, A = [[1, 2, 3], [0, -1, 1], [0, 0, 1]], confirming that f is both injective and surjective due to solvable equations relating to the rank and determinant properties of the matrix .
To find the inverse of a bijective function such as f(x) = 3 log2 x, we solve the equation y = 3 log2 x for x in terms of y. Rearranging gives y/3 = log2 x, leading to x = 2^(y/3) as the inverse function. This inverse ensures that applying f and then f^−1 returns the original input, confirming the bijection between the set of positive real numbers X and the range Y .
Discrete hash functions can be evaluated for efficiency in data storage based on how they manage collisions, which occur when multiple inputs map to the same hash value. The collision resolution policy is critical in determining efficiency. For instance, given a hash function h(x) = x mod 11 with cells indexed 0 to 10, and a sequence of data, the resolution policy is to occupy the next unoccupied cell when a collision occurs. The efficiency of this process can be seen by the specific placement of inputs such as 53, 13, 281, 743, 377, 20, 10, and 796 into initially empty cells following this policy, thereby maintaining a structured data storage despite collisions .
Decomposing a complex function into simpler component functions aids understanding and computation by isolating sub-processes or transformations. For example, the function f(x) = log2(x^2 + 2) can be decomposed into a series of simple functions: f1(z) = log2(z), f2(y) = y + 2, and f3(x) = x^2. Each component function performs a distinct part of the overall transformation, making analysis more manageable and facilitating easier computation for each step individually, enhancing clarity and reducing computational complexity .
The periodicity of a function like f(x) = x - ⌊x⌋, which is periodic with a period of 1, influences its graphical representation by repeating the same pattern across subsequent intervals. The function f(x) takes on values in the range [0, 1) in the interval [n, n+1) for any integer n, effectively duplicating the linear segment from 0 to just below 1 repeatedly along the x-axis at integer intervals. This periodic nature makes the function appear as a series of steps or spikes in the graph, with each step corresponding to a length of 1 on the x-axis .
The function f(n) = ⌈n/2⌉ is not injective because different inputs can produce the same output, which violates the one-to-one mapping requirement of injective functions. For instance, f(1) = ⌈1/2⌉ = 1 and f(2) = ⌈2/2⌉ = 1, hence 1 and 2 map to the same value. Despite this, the function is surjective since for every integer k, there exists an n such that f(n) = k, specifically by choosing n = 2k for non-negative integers .
The function that computes cardinality of subsets from a set X = {1, 2, ..., 10}, returning the size of each subset, is not injective because different subsets can have the same cardinality (e.g., {1} and {2} both have cardinality 1). It is not surjective because not all possible positive integers are achievable as cardinalities; specifically, numbers greater than 10 cannot be in the range. Thus, it lacks both injectivity and surjectivity despite representing a potentially large total number of subsets, evidenced by cardinalities ranging from 0 to 10, omitting integers outside this range .