Efficiently learning non-Markovianity of quantum dynamics with kernel-based quantum model

Chuan-Chi Huang^1*, Hong-Bin Chen^1,2,3

¹Department of Engineering Science, National Cheng Kung University, Tainan 70401, Taiwan
²Center for Quantum Frontiers of Research &Technology, NCKU, Tainan 701401, Taiwan
³Physics Division, National Center for Theoretical Sciences, Taipei 10617, Taiwan

* Presenter:Chuan-Chi Huang, email:k89208@gmail.com

In recent decades, the measure of non-Markovianity of open quantum systems has been widely
discussed. It is based on quantifying the flow of information between the open system and its
environment. Non-Markovianity could be seen as the reflow of information from the environment
to the system. To observe the reflow, we calculate the trace distance of the quantum states, after
which we can obtain the time evolution of the information flow. To get the time evolution, we have
to measure several times at different times. Our goal is to reduce the number of time points and
still get the exact non-Markovianity.

Keywords: Open quantum system, non-Markovianity , Machine learning