Skip to content
TheChatArchiveThe Chat Archive
o1-preview-2024-09-12· OpenAIAi2 WildChat-4.8M (AI2)

I'm trying to solve this question. What's the first hint in the proof you would give me.…

  1. User

    I'm trying to solve this question. What's the first hint in the proof you would give me. Do not solve the whole question. Suppose we have a vector space VV, in which we have chosen a basis e1,…,en\mathbf{e}_1, \dots, \mathbf{e}_n, giving rise to a dual basis α1,…,αn\alpha^1, \dots, \alpha^n for V∗V^*. As discussed in the lecture notes, this means that we can write any vector v∈V\mathbf{v} \in V as [ \mathbf{v} = [\mathbf{e}_1 \dots \mathbf{e}_n] \begin{bmatrix} v^1 \ \vdots \ v^n \end{bmatrix}, ] and vice versa any (dual) vector α∈V∗\alpha \in V^* as [ \alpha = [b_1 \dots b_n] \begin{bmatrix} \alpha^1 \ \vdots \ \alpha^n \end{bmatrix}. ]

    Here [v1⋮vn]\begin{bmatrix} v^1 \\ \vdots \\ v^n \end{bmatrix} and [b1…bn][b_1 \dots b_n] are matrices whose entries are scalars (real numbers), and [e1…en][\mathbf{e}_1 \dots \mathbf{e}_n] and [α1⋮αn]\begin{bmatrix} \alpha^1 \\ \vdots \\ \alpha^n \end{bmatrix} are matrices of vectors (with entries in respectively VV and V∗V^*).

    We now want to think about what happens if we change the basis for VV. Suppose we have another basis e~1,…,e~n\tilde{\mathbf{e}}_1, \dots, \tilde{\mathbf{e}}_n for VV, with corresponding dual basis α~1,…,α~n\tilde{\alpha}^1, \dots, \tilde{\alpha}^n for V∗V^*. With respect to these bases, we have v=[e~1…e~n][v~1⋮v~n],\mathbf{v} = [\tilde{\mathbf{e}}_1 \dots \tilde{\mathbf{e}}_n] \begin{bmatrix} \tilde{v}^1 \\ \vdots \\ \tilde{v}^n \end{bmatrix}, and α=[b~1…b~n][α~1⋮α~n].\alpha = [\tilde{b}_1 \dots \tilde{b}_n] \begin{bmatrix} \tilde{\alpha}^1 \\ \vdots \\ \tilde{\alpha}^n \end{bmatrix}.

    This gives rise to a change-of-basis matrix AA (an n×nn \times n-matrix with real coefficients) such that [ \begin{bmatrix} \tilde{v}^1 \ \vdots \ \tilde{v}^n \end{bmatrix} = A \begin{bmatrix} v^1 \ \vdots \ v^n \end{bmatrix}. ]

    In this set-up, how are [ [\tilde{\mathbf{e}}_1 \dots \tilde{\mathbf{e}}_n] \quad \text{and} \quad [\mathbf{e}_1 \dots \mathbf{e}_n], ] [ \begin{bmatrix} \tilde{\alpha}^1 \ \vdots \ \tilde{\alpha}^n \end{bmatrix} \quad \text{and} \quad \begin{bmatrix} \alpha^1 \ \vdots \ \alpha^n \end{bmatrix}, ] and [ [\tilde{b}_1 \dots \tilde{b}_n] \quad \text{and} \quad [b_1 \dots b_n] ] related to each other?

  2. Assistant

    To begin solving this problem, consider how the change of basis affects the basis vectors themselves. Since any vector (\mathbf{v}) can be expressed in terms of both the old basis ([\mathbf{e}_1 \dots \mathbf{e}_n]) and the new basis ([\tilde{\mathbf{e}}_1 \dots \tilde{\mathbf{e}}_n]), think about how these bases are related through the change-of-basis matrix (A). Specifically, try expressing the original basis vectors in terms of the new basis vectors using the matrix (A) or its inverse.