At half the size of Earth and one-tenth its mass, Mars is a featherweight as far as planets go. Yet new research reveals the extent to which Mars is quietly tugging on Earth’s orbit and shaping the cycles that drive long-term climate patterns here, including ice ages.
The study is published in the journal Publications of the Astronomical Society of the Pacific.
Stephen Kane, a professor of planetary astrophysics at UC Riverside, began this project with doubts about recent studies tying Earth’s ancient climate patterns to gravitational nudges from Mars. These studies suggest that sediment layers on the ocean floor reflect climate cycles influenced by the red planet despite its distance from Earth and small size.
For the first time, physicists in Italy have created a ‘lump soliton’: an extremely stable packet of light waves which can travel through 3D space, and even interact with other solitons without losing its shape.
Led by Ludovica Dieli at Sapienza University of Rome, the team achieved their result using a specially engineered crystal, whose responses to incoming light beams could be tightly controlled using an external voltage. Their study appears in Physical Review Letters.
Physicists from Trinity College Dublin believe new insights into the behavior of light may offer a new means of solving one of science’s oldest challenges—how to turn heat into useful energy.
Their theoretical leap forwards, which will now be tested in the lab, could influence the development of specialized devices that would ultimately increase the amount of energy we can capture from sunlight (and lamps and LEDs) and then repurpose to perform useful tasks.
The work has just been published in the journal, Physical Review A.
A team of physicists has discovered a surprisingly simple way to build nuclear clocks using tiny amounts of rare thorium. By electroplating thorium onto steel, they achieved the same results as years of work with delicate crystals — but far more efficiently. These clocks could be vastly more precise than current atomic clocks and work where GPS fails, from deep space to underwater submarines. The advance could transform navigation, communications, and fundamental physics research.
Thanks to their work from the 1980s and onward, John Hopfield and Geoffrey Hinton have helped lay the foundation for the machine learning revolution that started around 2010.
The development we are now witnessing has been made possible through access to the vast amounts of data that can be used to train networks, and through the enormous increase in computing power. Today’s artificial neural networks are often enormous and constructed from many layers. These are called deep neural networks and the way they are trained is called deep learning.
A quick glance at Hopfield’s article on associative memory, from 1982, provides some perspective on this development. In it, he used a network with 30 nodes. If all the nodes are connected to each other, there are 435 connections. The nodes have their values, the connections have different strengths and, in total, there are fewer than 500 parameters to keep track of. He also tried a network with 100 nodes, but this was too complicated, given the computer he was using at the time. We can compare this to the large language models of today, which are built as networks that can contain more than one trillion parameters (one million millions).
