Algorithm of Discovery – Defining What Isn’t
Jantzen: In my work, I’m making a computer do the job of looking for what doesn’t matter.
Is it possible to better understand what something is, by knowing what it isn’t? I’m Jim Metzner and this is the Pulse of the Planet.
By specifying all the things that don’t influence behavior, you are indirectly specifying all the kinds of behavior that’s possible.
Benjamin Jantzen is an assistant professor of philosophy in the College of Arts and Sciences at Virginia Tech. He’s devising algorithms sets of instructions for a computer designed to find new connections, new ideas in science.
Jantzen: This idea of of characterizing behavior by specifying what doesn’t matter for behavior, applies across the sciences: materials, chemistry, physics, biology.
The aim is to turn the way we typically characterize things upside down, and in the process make new discoveries.
Jantzen: By asking it to find us these sort of negative characterizations – the set of things that don’t matter, we’re looking to pick out new categories of things that do matter in ways that allow us to see new therapeutic relations between chemicals and health, right? New drugs, new ways in which material properties can relate. So we can get things like flexible electronics. New ways in which thermodynamic systems can relate. So you can get more efficient solar generation. Things of this sort – things that we missed because we’ve been focusing on an old positive picture.
It’s kinda like currently ask you to walk around the world and see the spaces people fill, rather than the people – to see the outlines of the shape rather than what’s inside the shape. .
You can hear this and previous programs on our podcast. I’m Jim Metzner and this is the Pulse of the Planet.