What's up in
Machine Learning Gets a Quantum Speedup
Two teams have shown how quantum approaches can solve problems faster than classical computers, bringing physics and computer science closer together.
Researchers Build AI That Builds AI
By using hypernetworks, researchers can now preemptively fine-tune artificial neural networks, saving some of the time and expense of training.
Any Single Galaxy Reveals the Composition of an Entire Universe
In computer simulations of possible universes, researchers have discovered that a neural network can infer the amount of matter in a whole universe by studying just one of its galaxies.
The Year in Math and Computer Science
Mathematicians and computer scientists answered major questions in topology, set theory and even physics, even as computers continued to grow more capable.
What Does It Mean for AI to Understand?
It’s simple enough for AI to seem to comprehend data, but devising a true test of a machine’s knowledge has proved difficult.
AI Researchers Fight Noise by Turning to Biology
Tiny amounts of artificial noise can fool neural networks, but not humans. Some researchers are looking to neuroscience for a fix.
The Uselessness of Useful Knowledge
Today’s powerful but little-understood artificial intelligence breakthroughs echo past examples of unexpected scientific progress.
Neuron Bursts Can Mimic Famous AI Learning Strategy
A new model of learning centers on bursts of neural activity that act as teaching signals — approximating backpropagation, the algorithm behind learning in AI.
A New Link to an Old Model Could Crack the Mystery of Deep Learning
To help them explain the shocking success of deep neural networks, researchers are turning to older but better-understood models of machine learning.