In Heisenberg's 1925 article Quantum Theoretical Interpretation of Kinematic and Mechanical Relations, one of the first things he establishes are the form of the frequency functions in (what I assume are) energy transitions: \begin{align*}& \nu(n,n-\alpha)=\frac{1}{h}[W(n)-W(n-\alpha)] \\ &\nu(n,\alpha)=\alpha\nu(n)=\frac{\alpha}{h} \frac{dW}{dn}\end{align*}
where the first relation corresponds to the quantum case and the second to the classical.
I think I recognize the first relation if $W$ is the energy of the energy levels (probably still reminiscent of Einsteins work function?). But, in that case, where does the classical expression come from? My knowledge on radiation theory is very limited, very basic stuff from an electrodynamics undergraduate course. If this is out of the scope of an answer I would be grateful if anyone could give me a reference instead.