Thanks all for the responses! (Sorry for the delayed upvotes and answer selection, I thought I did those already).
It was recently pointed out to me that there's a way to do this in a principled manner. Let's assume the lamp is 4W$4\ \mathrm W$ strong and distributes its power over 10 plates (1000 mm$^2$$1000\ \mathrm{mm^2}$). The energy of a 254nm$254\ \mathrm{nm}$ photon is $7.8 \times 10^{-19}$ Joules$7.8 \times 10^{-19}\ \mathrm J$. If the lamp is at 4W$4\ \mathrm W$, it's putting out 4 J/s$4\ \mathrm{J/s}$ of energy. This corresponds to $$ \frac{4 J}{s} * \frac{\mathrm{photon}}{7.8 \times 10^{-19} J} = 5.11 \times 10^{18}$$$$ \frac{4\ \mathrm J}{\mathrm s}\times\frac{\text{photon}}{7.8 \times 10^{-19}\ \mathrm J} = 5.11 \times 10^{18}$$ photons per second.
Now, if the absorbance of the initiator is 0.0001, we have that the absorbed fraction of the photons is: $$ \% Abs = 1 - \%T = 1 - \frac{I_1}{I_0} = 1 - 10^{-A} = 2 \times 10^{-4} $$
Since we have $5.11 \times 10^{18}$ photons per second are incoming, roughly $1 \times 10^{15}$ of them will be absorbed, which should be more than enough to initiate polymerization.
Of course, this is a very, very rough calculation, but I thought it was cool that you could, in theory, calculate that this experiment should work (which it did!)