a) (0.75) Sabendo \mathbf{T} é linear, determine \mathbf{T}\left(\mathbf{x}_{1}, \mathbf{x}_{2}, \mathbf{x}_{3}, \mathbf{x}_{4}, \mathbf{x}_{5}\right).
b) (0.75) Verificar que \mathbf{S} não é uma transformação linear usando a caracterização : \mathbf{S} \overrightarrow{\mathbf{v}}=\mathbf{A} \cdot \overrightarrow{\mathbf{v}}
c) (0.75) Verificar que \mathbf{S} não é uma transformação linear pela definição .
d) (0.75) Calcular o núcleo de \mathbf{T}: \mathbf{N u c}(\mathbf{T}).
e) (0.75) Calcular a imagem de \mathbf{T}: \mathbf{I m}(\mathbf{T}).
f) (0.75) Verificar se \mathbf{T} é injetiva .
g) (0.5) Verificar se \mathbf{T} é sobrejetiva .
h) (0.5) Determine \mathbf{T}^{-1}, se existir .
i) (1.0) Determinar uma base para \operatorname{Nuc}(\mathbf{T}) e \operatorname{Im}(\mathbf{T}).
j) (1.0) Determinar [\mathbf{T}]_{\beta}^{\alpha}, \operatorname{com} \alpha=\{(1,0,0,0,0),(0,2,0,0,0),(0,0,3,0,0),(0,0,0,4,0),(0,0,0,0,5)\} e \beta=\{(1,0,0,0,0),(0,1,0,0,0),(0,0,1,0,0),(0,0,0,1,0),(0,0,0,0,1)\}
Passo 1
O enunciado nos dá duas aplicações em espaços de dimensão 5 (e 4) e várias propriedades a verificar. O conteúdo principal é transformações lineares em \mathbb{R}^n:
Para \mathbf{T}:\mathbb{R}^5\to\mathbb{R}^5 precisamos usar linearidade, achar a fórmula geral, núcleo, imagem, testar injeção/sobrejeção, inversa e bases;
Para \mathbf{S}:\mathbb{R}^5\to\mathbb{R}^4 precisamos mostrar que não é linear, tanto pela não existência de matriz associada quanto pela falha nas propriedades de aditividade/homogeneidade.
Para haver \mathbf{S}(\mathbf{v})=\mathbf{A}\mathbf{v} toda coordenada de \mathbf{S} teria de ser uma combinação linear de (x,y,z,w,t). Mas a 2ª coordenada é \sqrt[3]{x^3+y^3+z^3+w^3},
que não é linear em x,y,z,w. Portanto não existe matriz \mathbf{A} tal que \mathbf{S}(\mathbf{v})=\mathbf{A}\mathbf{v}.
Passo 4
Pela definição, se \mathbf{S} fosse linear teríamos \mathbf{S}(v+w)=\mathbf{S}(v)+\mathbf{S}(w).
Escolha v=(1,0,0,0,0), w=(0,1,0,0,0).
Então
Como 2^{1/3}\neq2, falha aditividade, logo \mathbf{S} não é linear.
Passo 5
O núcleo de \mathbf{T}\) é a solução de \(\mathbf{T}(x)=0\), i.e.
\[\begin{cases} x_1+x_2+x_3+x_4+x_5=0,\\\ x_2+x_3+x_4+x_5=0,\\\ x_3+x_4+x_5=0,\\\ x_4+x_5=0,\\\ x_5=0. \end{cases}
Retro-substituindo dá x_5=0,x_4=0,x_3=0,x_2=0,x_1=0\].
Assim \(\ker(\mathbf{T})=\{0\}.
Passo 6
Como \dim\ker(\mathbf{T})=0 e \dim\mathbb{R}^5=5, por Teorema da Dimensão \dim\operatorname{Im}(\mathbf{T})=5. Logo \operatorname{Im}(\mathbf{T})=\mathbb{R}^5.
Isto implica que \mathbf{T} é injetiva (núcleo trivial) e, por terem domínios e codomínios de mesma dimensão, também é sobrejetiva.
Para achar \mathbf{T}^{-1}, resolvemos
Uma base de \ker(\mathbf{T}) é vazia (apenas o vetor zero).
Como \operatorname{Im}(\mathbf{T})=\mathbb{R}^5, uma base pode ser os vetores-padrão \{(1,0,0,0,0),\dots,(0,0,0,0,1)\}
ou as imagens \{\mathbf{T}(e_i)\}, que também são independente e geram.
Passo 8
Para \alpha=\{v_1,\dots,v_5\} com v_1=(1,0,0,0,0),v_2=(0,2,0,0,0),\dots,v_5=(0,0,0,0,5)
e \beta=\{e_1,\dots,e_5\}, as colunas de [\mathbf{T}]_{\beta}^{\alpha} são \mathbf{T}(v_i) escritas em \beta:
(a) \displaystyle \mathbf{T}(x_1,x_2,x_3,x_4,x_5)
=\bigl(x_1+x_2+x_3+x_4+x_5,\;x_2+x_3+x_4+x_5,\;x_3+x_4+x_5,\;x_4+x_5,\;x_5\bigr).
(b) \mathbf{S} não é linear pois a 2ª coordenada \sqrt[3]{x^3+y^3+z^3+w^3} não é combinação linear de (x,y,z,w,t).
(c) Pela definição, escolhendo v=(1,0,0,0,0), w=(0,1,0,0,0), obtém-se \mathbf{S}(v+w)\neq\mathbf{S}(v)+\mathbf{S}(w).
(d) \ker(\mathbf{T})=\{0\}.
(e) \operatorname{Im}(\mathbf{T})=\mathbb{R}^5.
(f) \mathbf{T} é injetiva (núcleo trivial).
(g) \mathbf{T} é sobrejetiva (dimensões iguais e imagem inteira).
(h) \displaystyle \mathbf{T}^{-1}(y_1,y_2,y_3,y_4,y_5)
=(y_1-y_2,\;y_2-y_3,\;y_3-y_4,\;y_4-y_5,\;y_5).
(i) Base de \ker(\mathbf{T}): \varnothing. Base de \operatorname{Im}(\mathbf{T}): \{(1,0,0,0,0),\dots,(0,0,0,0,1)\}.
(j) \displaystyle [\mathbf{T}]_{\beta}^{\alpha}
=\begin{pmatrix}
1&2&3&4&5\\\\ 0&2&3&4&5\\\\ 0&0&3&4&5\\\\ 0&0&0&4&5\\\\ 0&0&0&0&5
\end{pmatrix}.