Nautilus

The Fundamental Limits of Machine Learning

A few months ago, my aunt sent her colleagues an email with the subject, “Math Problem! What is the answer?” It contained a deceptively simple puzzle:

She thought her solution was obvious. Her colleagues, though, were sure their solution was correct—and the two didn’t match. Was the problem with one of their answers, or with the puzzle itself?

My aunt and her colleagues had stumbled across a fundamental problem in machine learning, the study of computers that learn. Almost all of the learning we expect our computers to do—and much of the learning we ourselves do —is about reducing information to underlying patterns, which can then be used to infer the unknown. Her puzzle was no different.

As a human, the challenge is to find any pattern at all. Of course, we have intuitions that

You’re reading a preview, subscribe to read more.

More from Nautilus

Nautilus8 min read
10 Brilliant Insights from Daniel Dennett
Daniel Dennett, who died in April at the age of 82, was a towering figure in the philosophy of mind. Known for his staunch physicalist stance, he argued that minds, like bodies, are the product of evolution. He believed that we are, in a sense, machi
Nautilus7 min readIntelligence (AI) & Semantics
How Whales Could Help Us Speak to Aliens
On Aug. 19, 2021, a humpback whale named Twain whupped back. Specifically, Twain made a series of humpback whale calls known as “whups” in response to playback recordings of whups from a boat of researchers off the coast of Alaska. The whale and the
Nautilus8 min read
A Revolution in Time
In the fall of 2020, I installed a municipal clock in Anchorage, Alaska. Although my clock was digital, it soon deviated from other timekeeping devices. Within a matter of days, the clock was hours ahead of the smartphones in people’s pockets. People

Related Books & Audiobooks