Home » Blog » Prove that given properties are still valid under an alternate definition of the dot product

Prove that given properties are still valid under an alternate definition of the dot product

Suppose that in \mathbb{R}^2 we use an alternate definition of the dot product,

    \[ (a_1, a_2) \cdot (b_1, b_2) = 2a_1 b_1 + a_2 b_2 + a_1 b_2 + a_2 b_1. \]

Prove that all of the properties of Theorem 12.2 (page 451 of Apostol) are still valid. Is the Cauchy-Schwarz inequality still valid?


Proof. Let A = (a_1, a_2), B = (b_1, b_2) and C = (c_1, c_2).

  1. We compute,

        \begin{align*}  A \cdot B &= 2a_1 b_1 + a_2 b_2 + a_1 b_2 + a_2 b_1 \\  &= 2 b_1 a_1 + b_2 a_2 + b_2 a_1 + b_1 a_2 \\  &= B \cdot A. \end{align*}

  2. We compute,

        \begin{align*}  A \cdot (B+C) &= A \cdot  ((b_1 + c_1, b_2 + c_2)) \\  &= 2a_1 (b_1 + c_1) + a_2 (b_2 + c_2) + a_1 (b_2 + c_2) + a_2 (b_1 + c_1) \\  &= 2a_1 b_1 + 2a_1 c_1 + a_2 b_2 + a_2 c_2 + a_1 b_2 + a_1 c_2 + a_2 b_1 + a_2 c_1 \\  &= 2a_1 b_1 + a_2 b_2 + a_1 b_2 + a_2 b_1 + 2a_1 c_1 + a_2 c_2 + a_1 c_2 + a_2 c_1 \\  &= A \cdot B + A \cdot C. \end{align*}

  3. We compute,

        \begin{align*}  c (A \cdot B) &= c (2a_1 b_1 + a_2 b_2 + a_1 b_2 + a_2 b_1) \\  &= 2(ca_1) b_1 + (ca_2)b_2 + (ca_1)b_2 + (ca_2) b_1 \\  &= (cA) \cdot B. \end{align*}

    On the other hand,

        \begin{align*}  c (A \cdot B) &= c (2a_1 b_1 + a_2 b_2 + a_1 b_2 + a_2 b_1) \\  &= 2a_1 (c b_1) + a_2 (cb_2) + a_1 (cb_2) + a_2 (cb_1) \\  &= A \cdot (cB).  \end{align*}

  4. We compute,

        \begin{align*}  A \cdot A &= 2a_1^2 + a_2^2 + a_1 a_2 + a_2 a_1 \\  &= a_1^2 + (a_1 + a_2)^2 \\  &> 0 \end{align*}

    if A \neq O.

  5. If A = O then

        \[ A \cdot A = 2a_1^2 + a_2^2 + 2a_1 a_2 = 0. \qquad \blacksquare \]

Cauchy-Schwarz still holds since we proved the Cauchy-Schwarz inequality using properties (a)-(e). Since these properties still hold, the same proof is valid.

Point out an error, ask a question, offer an alternative solution (to use Latex type [latexpage] at the top of your comment):