Deep learning from a dynamical viewpoint

NUS mathematicians have developed a new theoretical framework based on dynamical systems to understand when and how a deep neural network can learn arbitrary relationships.

Researchers reveal structure and function of a molecular motor

Molecular motors are complex devices composed of many different parts that consume energy to perform various cellular activities. In short, molecular machines transform energy into useful work. Understanding the mechanistical ...

Quantum systems and the flight of the bee

At first glance, a system consisting of 51 ions may appear easily manageable. But even if these charged atoms are only changed back and forth between two states, the result is more than two quadrillion (1015) different orderings ...

Dynamics of ocean worlds likely controlled by their rotation

Discovering that many of the large moons in the outer solar system may host significant subsurface oceans of liquid water has been a key advance in planetary science. These moons represent some of the most promising habitats ...

Reconstructing the states of a nonlinear dynamical system

We often encounter nonlinear dynamical systems that behave unpredictably, such as the Earth's climate and the stock market. To analyze them, measurements taken over time are used to reconstruct the state of the system. However, ...

page 3 from 14