The arXiv preprint server is home to cutting-edge scientific research, and academics often upload papers there before they're published. It's in this grand tradition that University of California San Diego postdoctoral scholar Eve Armstrong uploaded a paper called: "A Neural Networks Approach to Predicting How Things Might Have Turned Out Had I Mustered the Nerve to Ask Barry Cottonfield to the Junior Prom Back in 1997."
The paper is 10 pages long and extremely detailed, as Armstrong describes using advanced artificial intelligence algorithms known as neural networks (programs that "learn" how to make predictions based on a large amount of input data) to find out what Barry would have said had she asked him to the prom—basically, training a computer to think like Barry Cottonfield, high school heart throb, in 1997.
In case you couldn't tell already, the paper is dated April 1st, so it's a joke. Although, Armstrong wrote me in an email, it is inspired by real events. "That is my high school, and there was someone I never mustered the nerve to ask to the prom," she wrote.
And, since Armstrong is a real computer scientist, although the data and experiment are fake the methods is solid. "The algorithm is realistic, i.e. everything in Materials and Methods could be used to construct such work," she wrote.
So, how would Armstrong have done with Cottonfield, hypothetically speaking, way back in 1997? According to the paper:
Network performance on test data indicates that this author would have received an 87.2 (1)% chance of 'Yes' given a particular set of environmental input parameters. Most critically, the optimal method of question delivery is found to be Secret Note rather than Verbal Speech. There also exists mild evidence that wearing a burgundy mini-dress might have helped. The network performs comparably for all values of regularization strength, which suggests that the nature of noise in a high school hallway during passing time does not affect much of anything.
So much regret.