"""

Once upon a time—I’ve seen this story in several versions and several places, sometimes cited as fact, but I’ve never tracked down an original source—once upon a time, I say, the US Army wanted to use neural networks to automatically detect camouflaged enemy tanks.

The researchers trained a neural net on 50 photos of camouflaged tanks amid trees, and 50 photos of trees without tanks. Using standard techniques for supervised learning, the researchers trained the neural network to a weighting that correctly loaded the training set—output “yes” for the 50 photos of camouflaged tanks, and output “no” for the 50 photos of forest.

Now this did not prove, or even imply, that new examples would be classified correctly. The neural network might have “learned” 100 special cases that wouldn’t generalize to new problems. Not, “camouflaged tanks versus forest”, but just, “photo-1 positive, photo-2 negative, photo-3 negative, photo-4 positive…” But wisely, the researchers had originally taken 200 photos, 100 photos of tanks and 100 photos of trees, and had used only half in the training set. The researchers ran the neural network on the remaining 100 photos, and without further training the neural network classified all remaining photos correctly. Success confirmed!

The researchers handed the finished work to the Pentagon, which soon handed it back, complaining that in their own tests the neural network did no better than chance at discriminating photos. It turned out that in the researchers’ data set, photos of camouflaged tanks had been taken on cloudy days, while photos of plain forest had been taken on sunny days. The neural network had learned to distinguish cloudy days from sunny days, instead of distinguishing camouflaged tanks from empty forest.

"""

That is a pretty novice mistake for a paid scientist, imo.

The story apparently dates from 1963. I wouldn't judge any scientist training neural networks in the 60s too harshly.

I certainly did not expect the down votes. I've designed a number psychology experiments and I and my colleagues were always careful to balance experimental stimuli and conditions on a number of factors, ranging from luminosity to semantic association, and especially order.

When I was tasked to create stimuli or to take measurements on a series of items it was always important to try to eliminate systematic differences on non-interest, most typically thru randomization of the order of creation or measurements.

I would point out that the study which appears to be the closest thing to a tiny seed of truth to the tank story (Kanal & Randall 1964) actually did balance its photographs by construction, by subcropping large aerial photographs to have tank/no-tank sections and training on that. That necessarily controls for time-of-day, weather, luminosity etc.

apocryphal ehh?