Video: Copying vs. Transforming Information

New research by SFI Postdoctoral Fellow Artemy Kolchinsky and Bernat Corominas-Murtra presents an important distinction for information theory — copying vs. transforming. Watch the video explainer.

Read More

Learning by omission

What would happen if neural networks were explicitly trained to discard useless information, and how to tell them to do so, is the subject of recent research by SFI's Artemy Kolchinsky, Brendan Tracey, and David Wolpert.

Read More

Biological and physical time meet in sleep

A working group, held November 18-20 at SFI, is beginning to unpack the causes, timescales, and consequences of sleep. In particular, participants are focusing on how sleep time changes across species, and how it changes with age and during adulthood.

Read More

Can evolution reveal how life emerged from chemistry?

A group of biologists think that a new synthesis in evolutionary theory might help answer the question of how life’s progenitor originally emerged. A working group, meeting November 13-15, brings together evolutionary theorists and experimentalists to explore which evolutionary models might best explain how chemical systems become biological systems.

Read More

New Book: The Ethical Algorithm

In The Ethical Algorithm: The Science of Socially Aware Algorithm Design, SFI External Professor Michael Kearns and his University of Pennsylvania colleague Aaron Roth offer a set of principled solutions based on the emerging science of socially aware algorithm design.

Read More

Information theory as a tool for extracting climate signals

During Earth’s last glacial period, temperatures on the planet periodically spiked dramatically and rapidly. A new paper in the journal Chaos by SFI's Joshua Garland, Liz Bradley, and coauthors suggests that mathematics from information theory could offer a powerful tool for analyzing and understanding them.

Read More

Looking for entrenchment in all the right places

Ashley Teufel and Luis Zaman's working group, “The Point of No Return,” seeks to identify the underlying properties driving entrenchment, a phenomenon in which a single event can have a widespread effect on an entire system, and find ways to infer, predict or even control it.

Read More