Which alloy and temper? You have to overcome the yield strength to get the part back to shape. Laying weight on it will only waste your time. The part has to be bent in the opposite direction of the bow, but only enough to get past the yield strength. If you're not familiar with yield strength. The two most important mechanical properties in aluminum is the Yield strength and the Ultimate Tensile Strength. When aluminum is deformed, it will spring back, up to a point. Aluminum will reach a point of stress where it will not return to it's previous shape, it "yields". After aluminum yields, it will stretch a certain amount, depending on the alloy/temper, and then it will break. The ultimate strength is a measure of the force it took to break the sample. Well there is another property called elongation, it is basically the length the part stretched before it broke, think final length subtracting the original length, as a percentage of the original length. It's the "distance" between the Yield and the Ultimate. We deal with 3xxx and 6xxx alloys. 3003 O-Temp will elongate up to 60% before it breaks, where a 6061 T9 might only stretch 5% before it breaks.
If you have a 6xxx alloy, depending on the temper, the yield strength could be anywhere from 9,000 lbs to 45,000 lbs.
If it's a 3xxx series (unlikely), the yield would be somewhere between 5,000 lbs to 25,000 lbs, again, depending on the temper.
You could try heating it up. The yield strength drops considerably with temperature. Heat just the the area of the bow with a torch, you're going to need hot mitts, then carefully pull against the bow, don't over-do, if you kink it, it's done. Don't get too much of the part hot, or you end up with multiple bows. I wouldn't go over 400f, you'll blow up the grain and end up with a brittle part.
another edit, a poster below had a good idea with heating up the part. You might try heating the concave side of the bow, pretty hot, but you need to cool it off quickly. Aluminum shrinks as it cools. We see this in extrusion, the parts come out at 1,000f and if one side of the part is thinner than the other, that side will cool, and shrink faster, causing our extrusions to bow. But we have stretchers to remove it.