Home About Us Research People Visitors International Cooperation News Education & Training Join Us Papers Resources Links Societies & Publications
Int’l Cooperation News
Pic News
Forums & Lectute
Location: Home > News > Events
Why does classifying images appear to be easier than simulating two-dimensional quantum systems?
Date: 2018-07-09
Time: 11:00
Venue: M830
Speaker: Yi-Chen Huang


Abstract: I will argue that classifying images is significantly easier than simulating two-dimensional quantum systems from the perspective of entanglement scaling. In particular, I show that under reasonable assumptions, the entanglement between a region and its complement scales as the logarithm of the boundary length for image classification problems. This implies a provably efficient neural network representation for the function that maps an image to the label the image corresponds to.
Reference: arXiv:1711.04606

Contact: Lei Wang 9853

The Institute of Physics, Chinese Academy of Sciences P.O.Box 603,Beijing 100190,China
Tel:86-10-82649361 Fax:86-10-82649531 email: zhoumb@aphy.iphy.ac.cn
© Copyright The Institute of Physics