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

Nautilus3 min read
Sardines Are Feeling the Squeeze
Sardines are never solitary. Even in death they are squeezed into a can, three or five to a tin, their flattened forms perfectly parallel. This slick congruity makes sense. In life, sardines are evolved for synchronicity: To avoid and confuse predato
Nautilus6 min readIntelligence (AI) & Semantics
Why AI Can Never Make Humans Obsolete
This article is part of series of Nautilus interviews with artists, you can read the rest here. Angie Wang is a Los Angeles-based artist who has thought a lot about AI, and even more about what it means to be a human. Her illustrated essay for The Ne
Nautilus9 min read
Discovering the First Other Earths
Yes, you can get bad coffee in Vienna. Vienna is known for its beautiful cafés, where philosophers, poets, and scientists have found inspiration over endless cups of fantastic coffee for hundreds of years. Coffee is still at the heart of Viennese cul

Related