How did a computer scientist such as Geoffrey Hinton manage to win a Nobel Prize in physics when computer science already has its own Nobel Prize equivalent in the Turing Awards?
Last Updated: 23.06.2025 01:38
"Didn't he do something with Boltzmann in it? That sounds physics-y. RBMs and stuff, eh?"
You can have your cake and eat it too
"Good point, I'm sure we can swing it. And let's tack on Hopfield while we're about it."
Fortunately, we are privy to the discussion that led up to this:
… then anything is possible. There’s no rule that a Nobel Prize has to make sense.^*
^† They rationalise their decision thusly:
"Hey guys, AI is pretty big so let's centre our prizes on it this year. We can get some attention, and it's all about advertising, at the end of the day, isn't it?"
Whatever.
When he's standing, in front of you
Why does my vagina always itch so badly after my periods?
Why wait any longer for the one you love?
[Younger voice] "But wait a minute, Ising-Lenz goes back to the 1920's. And didn't Hinton plagiarise rather a lot? He also didn't invent modern backprop, did he, that's Linnainmaa? And Amari preceded Hopfield, too. That's not a good look."
(Bob Dylan, Nobel Prize for Literature, 2016)
A fly on the wall at the Royal Swedish Academy of Sciences
In awarding prizes, the Nobel Committees often seem only marginally more competent than MTG is at explaining meteorology. And if they can give a literature prize for lyrics like:
"Good idea, but how can we wangle something that says 'Physics'?"
Why wait any longer for the world to begin?
"Naah, Linnainmaa is a Finn. Can't give it to a bloody Finnish mathematician. Let's go for drinks. Brännvin anyone?"
^* Fibiger got the 1926 Medicine prize for the discovery of Spiroptera carcinoma (Don’t ask).
Gucci Owner Picks Auto Executive for One of Global Luxury’s Top Jobs - WSJ
My 11 million SEK, Dr Jo.
In December 1973, when Kissinger was awarded the Nobel Peace Prize, comedian Tom Lehrer dropped his mic and stamped on it—satire had just died.
There you go.^†
They then move on to selectively provide their own version of history. But hey, it’s OK. They wanted controversy, didn’t they? Whatever.
[Older voice] "Mmm. What about Hinton, he's widely regarded? Nobody got fired for buying IBM"
(Mumbles of assent)
Whatever.
[The basic structure of artificial neural networks] has close similarities with spin models in statistical physics applied to magnetism or alloy theory. This year’s Nobel Prize in Physics recognizes research exploiting this connection to make breakthrough methodological advances in the field of ANN.
"Where can we shoehorn it in? Chemistry is easy 'cos AlphaFold; but what about physics? A bit more challenging, right?"