I apologize for the long post in advance, but I got as in depth as possible to as the question raises some issues that are worth taking the time to understand.
The problem is largely that there's no easy single answer to your question.
For one, the temperature differential is affected by a number of factors. Although you could say that the level of yeast activity is pretty much THE determining factor, THIS factor is in turn determined by many other factors, including variables such as the particular strain(s) of yeast used, wort composition, ambient temperature, pitching rates, head pressure, and even fermentor geometry.
But that's only a small part of the problem, really. Because even if you understood all these variables well enough to predict how they would affect the temperature differential as a whole... even if you you could somehow reliably predict how high above ambient it's going to go, within 0.1 degrees (which is just absolutely impossible), you're ignoring the biggest factor of all, a variable that you can't simply just figure out and use to determine the temp differential, because whether you like it or not, it's constantly changing - that factor is TIME.
And why is time such an issue? Because the yeast reproduces and attenuates over time, meaning the level of yeast activity is CONSTANTLY changing while the wort is still fermenting (which is precisely when temperature control is so particularly important). And what this results in is a temperature differential that is ALSO constantly changing.
So let's say the fermentation temp is going to eventually reach about 8° over ambient, which is pretty typical in a homebrewing setup, and, for the sake of simplicity, that you pitched the yeast with the wort at the same temp as ambient. Over the span of the fermentation, the temperature differential is going to increase over a period of time, until the yeast activity hits full swing. At this point, it will at 8° above ambient (just in this example). Then once the yeast has attenuated most of the sugars and begins to slowly work on the more complex sugars, the yeast activity dies down a, and the temperature differential will start to disappear as the fermentor cools down a bit.
This is especially an issue, because the temperature is decreasing at precisely the time that the yeast could most use an INCREASE in temps, as they struggle to ferment the last and most complex sugars in the wort, and could even in the worst case cause fermentation to stall. But far more commonly, the beer ends up not attenuating as well as it could, often finishing a few gravity points higher than if the wort temperature remained constant (or even increased a little towards the end to help the yeast attenuate fully.)
In other words, trying to control the fermentation temperature by adjusting the ambient temperature to account for the extra heat generated by the yeast is pointless, because even if you WERE somehow able to reasonably predict the CONSTANTLY changing temperature differential at any given time (which is already a totally unrealistic idea), you would need to be ALSO CONSTANTLY changing the ambient temp in order to keep up with it, just to maintain a relatively stable ferm temp.
So basically, you can't REALLY control ferm temps merely by taking a differential into account. All you're really accomplishing is the same thing you'd be able to accomplish by fermenting in a slightly cooler room. And that's still far from ideal because, as I mentioned earlier, it allows the temperature to drop at the WORST possible time. So if you're going to invest in any sort of temperature control (and I would recommend every brewer do so, regardless of experience), you really should just do it right, and DIRECTLY control the actual fermentation temperature - either by buying a temperature controller and either using a thermowell or simply insulating the temp probe against the fermentor wall, or even with a simple swamp cooler, if you're willing to sacrifice precision in order to save a bit of money. I highly recommend an electronic temp controller though, ESPECIALLY since you already have a fridge to ferment in... for a mere $20, you can significantly improve your temp control.
