{"id":8164,"date":"2021-10-04T18:45:29","date_gmt":"2021-10-04T18:45:29","guid":{"rendered":"http:\/\/TheNextWeb=1368842"},"modified":"2021-10-04T18:45:29","modified_gmt":"2021-10-04T18:45:29","slug":"codifying-humanity-can-humor-be-reduced-to-an-algorithm","status":"publish","type":"post","link":"https:\/\/www.londonchiropracter.com\/?p=8164","title":{"rendered":"Codifying humanity: Can humor be reduced to an algorithm?"},"content":{"rendered":"\n

Welcome to \u201cCodifying Humanity.\u201d A new Neural series that analyzes the machine learning world\u2019s attempts at creating human-level AI. <\/em><\/p>\n

Stop me if you\u2019ve heard this one. A robot walks into a bar and the bartender takes its order. The robot says: \u201cI\u2019ll have whatever my developer likes.\u201d<\/p>\n

If you\u2019re not laughing right now it\u2019s because the joke isn\u2019t funny. And if you are laughing, it\u2019s because the joke is <\/i>funny. That\u2019s how jokes work. It\u2019s also how people work. <\/span><\/p>\n

Humorous or not, the premise of the joke is that robots don\u2019t have personalities, ideas, thoughts, or desires. Any human-like qualities we could attribute to a machine or its output are merely reflections of ourselves or its programmers. <\/span><\/p>\n

That doesn\u2019t sit well with the mainstream perceptions of AI. We\u2019ve seen a hundred or more variations on the \u201cA robot wrote this article\u201d trope that <\/span>The Guardian<\/i> got caught up in last year<\/a>. Each one promises a near-future where human creators are either displaced or forced to work in tandem with machines.<\/span><\/p>\n

The common refrain is that AI isn\u2019t human-like yet, but <\/span>it will be sooner than you think!<\/i><\/p>\n

And, maybe after reading some carefully-curated outputs from OpenAI\u2019s text generator, GPT-3, it starts to sound less like hyperbole and more like good old common sense.<\/span><\/p>\n

We see back-flipping robots doing parkour and deepfake face-swaps in our social media feeds everyday. We have every reason to believe a text-generator can do things that seem straight out of the realm of science fiction.<\/span><\/p>\n

At least, until we start picking at the seams. Because, unfortunately, a functional understanding of the machinations of deep learning-based AI systems doesn\u2019t fall within the realm of common sense. <\/span><\/p>\n

Prestidigitation<\/span><\/h2>\n

Here at Neural, we refer to most of what AI does as prestidigitation. That\u2019s because there\u2019s only a handful of things a typical deep learning system can actually do. Much like a real-world magician, developers create incredible programs out of some fairly basic algorithmic foundations.<\/span><\/p>\n

The only difference between a disappearing coin trick and what David Copperfield does is <\/span>scale. <\/i>There is no more or less \u201creal magic\u201d involved in the former\u2019s illusions and the latter\u2019s. <\/span><\/p>\n

And it\u2019s the same with AI. Tesla\u2019s computer vision systems are no more or less human-like than Not Hotdog\u2019s<\/a>. They essentially perform the exact same function at different scales. <\/span><\/p>\n

It\u2019s hard to explain the simplicity of a massively complex AI system to the average person.<\/span><\/p>\n

So let\u2019s take something uniquely human and break down exactly what happens when you try to codify it for machines in the simplest possible way.<\/span><\/p>\n

Can an AI be funny? <\/span><\/h2>\n

Luckily for us, a former Microsoft intern named Nabil Hossain has already done all the groundwork for us. A few years back, Hossain and a pair of Microsoft AI researchers developed a machine learning system to generate humorous headlines from existing news articles<\/a>.<\/span><\/p>\n

The big idea was that the AI would make microedits by changing a single word in a serious headline to make it a funny one. <\/span><\/p>\n

\"A
<\/i><\/a>Do you think these are funny?<\/figcaption>