In quantum mechanics, the state of a system isn't described by position and velocity, but by a state vector. This vector doesn't live in the familiar 3D space of our everyday experience. Instead, it resides in a mathematical realm called a Hilbert space—an abstract vector space that provides the complete stage for quantum phenomena. For our purposes, you can think of it as a generalization of the Euclidean space you're used to, but where vectors can have complex numbers as components and can even have an infinite number of dimensions.

Just like we can calculate the dot product of two vectors to find the projection of one onto another, Hilbert spaces have a similar tool called an inner product. The inner product of two state vectors tells us how much they overlap. If the inner product is zero, the states are orthogonal, meaning they are completely distinct possibilities for the system. For example, the states corresponding to an electron's spin being 'up' and 'down' are orthogonal.

Writing out column vectors and row vectors can get cumbersome, especially in quantum calculations. To streamline this, physicists use Dirac notation, also known as bra-ket notation. A quantum state vector is represented by a 'ket', like $|ψ⟩$. This is conceptually equivalent to a column vector.

The corresponding 'bra', written $〈ψ|$, is the conjugate transpose of the ket. Think of it as a row vector where each component has been replaced by its complex conjugate.

Why the split? This notation makes the inner product incredibly elegant. The inner product of two states $|ψ⟩$ and $|φ⟩$ is written simply as $〈φ|ψ⟩$. This 'bra-ket' forms a single complex number, representing the projection of $|ψ⟩$ onto $|φ⟩$. The square of its absolute value, $|〈φ|ψ⟩|^2$, gives the probability of finding the system in state $|φ⟩$ if it was initially in state $|ψ⟩$.

In quantum mechanics, physical properties you can measure—like position, momentum, or energy—are called observables. Each observable is represented by a specific type of matrix called a Hermitian operator (often denoted with a 'hat', like $\hat{A}$). The key property of a Hermitian operator is that it is equal to its own conjugate transpose, $\hat{A} = \hat{A}^\dagger$.

This might seem like a purely mathematical detail, but it has a profound physical consequence: the eigenvalues of a Hermitian operator are always real numbers. This is a requirement, since the result of a physical measurement (like 3.4 joules of energy) must be a real number, not a complex one.

When we measure an observable $\hat{A}$ on a system in state $|ψ⟩$, two things happen:

1. The result of the measurement will be one of the eigenvalues of $\hat{A}$.
2. The state of the system instantly collapses into the eigenstate corresponding to that eigenvalue.

So, how do we predict the average outcome if we were to perform the same measurement on many identically prepared systems? This is the expectation value, written as $〈ψ|\hat{A}|ψ⟩$. It's a weighted average of the eigenvalues, where the weights are the probabilities of measuring each outcome.

The act of measurement can be formalized using projection operators. For a given eigenstate $|a_i⟩$ of an operator $\hat{A}$, the operator $\hat{P}_i = |a_i⟩〈a_i|$ is a projection operator. When it acts on a general state $|ψ⟩$, it 'projects' $|ψ⟩$ onto the direction of $|a_i⟩$.

A crucial property of the set of all eigenstates of a Hermitian operator is that they form a complete basis for the Hilbert space. This means any state vector can be written as a linear combination of these eigenstates. Mathematically, this is expressed by the completeness relation, which states that the sum of all projection operators for a given basis is the identity operator, $\hat{I}$.

This relation is incredibly useful. For instance, you can insert the identity operator anywhere in an expression without changing it. By inserting $\sum_i |a_i⟩〈a_i|$, you can decompose a state or an operation in terms of a convenient basis, which is a common trick for solving problems in quantum mechanics.