aboutsummaryrefslogtreecommitdiffstats
path: root/doc/exercises.tex
blob: d170bb988cf3c3a6dcc4208270094e8a4e83a3c3 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
\chapter{Exercises}
\label{chap:Exercises}

% TODO
% 
% All exercises should be group with a chapter
% 
% ▶   Recommended
% M   Matematically oriented
% HM  Higher matematical education required
% 
% 00  Immediate
% 10  Simple
% 20  Medium
% 30  Moderately hard
% 40  Term project
% 50  Research project
% 
% ⁺   High risk of undershoot difficulty


\begin{enumerate}[label=\textbf{\arabic*}.]


\item {[\textit{M10}]} \textbf{Convergence of the Lucas Number ratios}

Find an approximation for $\displaystyle{ \lim_{n \to \infty} \frac{L_{n + 1}}{L_n}}$,
where $L_n$ is the $n^{\text{th}}$ Lucas Number \psecref{sec:Lucas numbers}.

\( \displaystyle{
    L_n \stackrel{\text{\tiny{def}}}{\text{=}} \left \{ \begin{array}{ll}
      2 & \text{if} ~ n = 0 \\
      1 & \text{if} ~ n = 1 \\
      L_{n - 1} + L_{n + 1} & \text{otherwise}
    \end{array} \right .
}\)



\item {[\textit{M12${}^+$}]} \textbf{Factorisation of factorials}

Implement the function

\vspace{-1em}
\begin{alltt}
   void factor_fact(z_t n);
\end{alltt}
\vspace{-1em}

\noindent
which prints the prime factorisation of $n!$ (the $n^{\text{th}}$ factorial).
The function shall be efficient for all $n$ where all primes $p \le n$ can
be found efficiently. You can assume that $n \ge 2$. You should not evaluate $n!$.



\item {[\textit{M20}]} \textbf{Reverse factorisation of factorials}

You should already have solved ``Factorisation of factorials''
before you solve this problem.

Implement the function

\vspace{-1em}
\begin{alltt}
   void unfactor_fact(z_t x, z_t *P,
        unsigned long long int *K, size_t n);
\end{alltt}
\vspace{-1em}

\noindent
which given the factorsation of $x!$ determines $x$.
The factorisation of $x!$ is
$\displaystyle{\prod_{i = 1}^{n} P_i^{K_i}}$, where
$P_i$ is \texttt{P[i - 1]} and $K_i$ is \texttt{K[i - 1]}.



\item {[\textit{05}]} \textbf{Fast primality test}

$(x + y)^p \equiv x^p + y^p ~(\text{Mod}~p)$
for all primes $p$ and for a few composites $p$.
Use this to implement a fast primality tester.



\end{enumerate}



\chapter{Solutions}
\label{chap:Solutions}


\begin{enumerate}[label=\textbf{\arabic*}.]

\item \textbf{Convergence of the Lucas Number ratios}

It would be a mistake to use bignum, and bigint in particular,
to solve this problem. Good old mathematics is a much better solution.

$$ \lim_{n \to \infty} \frac{L_{n + 1}}{L_n} = \lim_{n \to \infty} \frac{L_{n}}{L_{n - 1}} = \lim_{n \to \infty} \frac{L_{n - 1}}{L_{n - 2}} $$

$$ \frac{L_{n}}{L_{n - 1}} = \frac{L_{n - 1}}{L_{n - 2}} $$

$$ \frac{L_{n - 1} + L_{n - 2}}{L_{n - 1}} = \frac{L_{n - 1}}{L_{n - 2}} $$

$$ 1 + \frac{L_{n - 2}}{L_{n - 1}} = \frac{L_{n - 1}}{L_{n - 2}} $$

$$ 1 + \varphi = \frac{1}{\varphi} $$

So the ratio tends toward the golden ratio.



\item \textbf{Factorisation of factorials}

Base your implementation on

\( \displaystyle{
    n! = \prod_{p~\in~\textbf{P}}^{n} p^{k_p}, ~\text{where}~
    k_p = \sum_{a = 1}^{\lfloor \log_p n \rfloor} \lfloor np^{-a} \rfloor.
}\)

There is no need to calculate $\lfloor \log_p n \rfloor$,
you will see when $p^a > n$.



\item \textbf{Reverse factorisation of factorials}

$\displaystyle{x = \max_{p ~\in~ P} ~ p \cdot f(p, k_p)}$,
where $k_p$ is the power of $p$ in the factorisation
of $x!$. $f(p, k)$ is defined as:

\vspace{1em}
\hspace{-2.8ex}
\begin{minipage}{\linewidth}
\begin{algorithmic}
    \STATE $k^\prime \gets 0$
    \WHILE{$k > 0$}
      \STATE $a \gets 0$
      \WHILE{$p^a \le k$}
        \STATE $k \gets k - p^a$
        \STATE $a \gets a + 1$
      \ENDWHILE
      \STATE $k^\prime \gets k^\prime + p^{a - 1}$
    \ENDWHILE
    \RETURN $k^\prime$
\end{algorithmic}
\end{minipage}
\vspace{1em}



\item \textbf{Fast primality test}

If we select $x = y = 1$ we get $2^p \equiv 2 ~(\text{Mod}~p)$. This gives us

\vspace{-1em}
\begin{alltt}
enum zprimality ptest_fast(z_t p)
\{
    z_t a;
    int c = zcmpu(p, 2);
    if (c <= 0)
      return c ? NONPRIME : PRIME;
    zinit(a);
    zsetu(a, 1);
    zlsh(a, a, p);
    zmod(a, a, p);
    c = zcmpu(a, 2);
    zfree(a);
    return c ? NONPRIME : PROBABLY_PRIME;
\}
\end{alltt}



\end{enumerate}