What's up in
Researchers Discover a More Flexible Approach to Machine Learning
“Liquid” neural nets, based on a worm’s nervous system, can transform their underlying algorithms on the fly, giving them unprecedented speed and adaptability.
Machines Learn Better if We Teach Them the Basics
A wave of research improves reinforcement learning algorithms by pre-training them as if they were human.
The Computer Scientist Who Finds Life Lessons in Games
In Shang-Hua Teng’s work, theoretical and practical questions have long been intertwined. Now he’s turning his focus to the impractical.
Finally, a Fast Algorithm for Shortest Paths on Negative Graphs
Researchers can now find the shortest route through a network nearly as fast as theoretically possible, even when some steps can cancel out others.
New Algorithm Closes Quantum Supremacy Window
Random circuit sampling, a popular technique for showing the power of quantum computers, doesn’t scale up if errors go unchecked.
The Physics Principle That Inspired Modern AI Art
Diffusion models generate incredible images by learning to reverse the process that, among other things, causes ink to spread through water.
The Year in Computer Science
Computer scientists this year learned how to transmit perfect secrets, why transformers seem so good at everything, and how to improve on decades-old algorithms (with a little help from AI).
What Does It Mean to Align AI With Human Values?
Making sure our machines understand the intent behind our instructions is an important problem that requires understanding intelligence itself.
After a Quantum Clobbering, One Approach Survives Unscathed
A quantum approach to data analysis that relies on the study of shapes will likely remain an example of a quantum advantage — albeit for increasingly unlikely scenarios.