7. For any vectors $\vec { a } , \vec { b }$, which of the following relations always holds?
A. $| \vec { a } \bullet \vec { b } | \leq | \vec { a } | | \vec { b } |$
B. $| \vec { a } - \vec { b } | \leq | | \vec { a } | - | \vec { b } | |$
C. $( \vec { a } + \vec { b } ) ^ { 2 } = | \vec { a } + \vec { b } | ^ { 2 }$
D. $( \vec { a } + \vec { b } ) ( \vec { a } - \vec { b } ) = \vec { a } ^ { 2 } - \vec { b } ^ { 2 }$

Let $\lambda$ be a real such that $0 \leqslant \lambda \leqslant 1$. For $t$ in $\{-1, 1\}$, $Y_{t}$ denotes the projection of $X'$ onto $C_{t}$. Show that
$$d(X, C) \leqslant \left\|(1 - \lambda)(Y_{\varepsilon_{n}} + \varepsilon_{n} e_{n}) + \lambda(Y_{-\varepsilon_{n}} - \varepsilon_{n} e_{n}) - X\right\|$$

Let $\lambda$ be a real such that $0 \leqslant \lambda \leqslant 1$. For $t$ in $\{-1, 1\}$, $Y_{t}$ denotes the projection of $X'$ onto $C_{t}$. Deduce that
$$d(X, C)^{2} \leqslant 4\lambda^{2} + \left\|(1 - \lambda)(Y_{\varepsilon_{n}} - X') + \lambda(Y_{-\varepsilon_{n}} - X')\right\|^{2}$$
then that
$$d(X, C)^{2} \leqslant 4\lambda^{2} + (1 - \lambda)\|Y_{\varepsilon_{n}} - X'\|^{2} + \lambda\|Y_{-\varepsilon_{n}} - X'\|^{2}$$
Thus, show the inequality
$$d(X, C)^{2} \leqslant 4\lambda^{2} + (1 - \lambda)d(X', C_{\varepsilon_{n}})^{2} + \lambda d(X', C_{-\varepsilon_{n}})^{2}$$

Let $n$ be an integer such that $n \geqslant 2$. We denote by $E' = \operatorname{Vect}(e_{1}, \ldots, e_{n-1})$ and by $\pi$ the orthogonal projection onto $E'$. We set $X' = \pi \circ X = \sum_{i=1}^{n-1} \varepsilon_{i} e_{i}$. For $t$ in $\{-1, 1\}$ we denote $C_{t} = \pi(C \cap H_{t})$ where $H_t = E' + te_n$. For $t$ in $\{-1, 1\}$, we denote by $Y_{t}$ the projection of $X'$ onto the non-empty closed convex set $C_{t}$. Let $\lambda$ be a real such that $0 \leqslant \lambda \leqslant 1$.
Show that
$$d(X, C) \leqslant \left\|(1 - \lambda)(Y_{\varepsilon_{n}} + \varepsilon_{n} e_{n}) + \lambda(Y_{-\varepsilon_{n}} - \varepsilon_{n} e_{n}) - X\right\|$$

Let $n$ be an integer such that $n \geqslant 2$. We denote by $E' = \operatorname{Vect}(e_{1}, \ldots, e_{n-1})$ and by $\pi$ the orthogonal projection onto $E'$. We set $X' = \pi \circ X = \sum_{i=1}^{n-1} \varepsilon_{i} e_{i}$. For $t$ in $\{-1, 1\}$ we denote $C_{t} = \pi(C \cap H_{t})$ where $H_t = E' + te_n$. For $t$ in $\{-1, 1\}$, we denote by $Y_{t}$ the projection of $X'$ onto the non-empty closed convex set $C_{t}$. Let $\lambda$ be a real such that $0 \leqslant \lambda \leqslant 1$.
Deduce that
$$d(X, C)^{2} \leqslant 4\lambda^{2} + \left\|(1 - \lambda)(Y_{\varepsilon_{n}} - X') + \lambda(Y_{-\varepsilon_{n}} - X')\right\|^{2}$$
then that
$$d(X, C)^{2} \leqslant 4\lambda^{2} + (1 - \lambda) \|Y_{\varepsilon_{n}} - X'\|^{2} + \lambda \|Y_{-\varepsilon_{n}} - X'\|^{2}$$
Thus, we have shown the inequality
$$d(X, C)^{2} \leqslant 4\lambda^{2} + (1 - \lambda) d(X', C_{\varepsilon_{n}})^{2} + \lambda d(X', C_{-\varepsilon_{n}})^{2}$$