media archive
Physics Conferences 7th KIAS Conference on Statistical Physics
Loading the player ...
[Talk 9] Reconstructing Network connectivity and weights from Noisy Dynamical Data
SPEAKER  |  Pik-Yin Lai
INSTITUTE  |  National Central University
DATE  |  July 5(Tue), 2016
TIME  |  09:40
PLACE  |  Korea Institute for Advanced Study (KIAS), Seoul, Korea
Keyword  |  
Download  |  
ABSTRACT  |  At the merge of the "Big data" era with huge amount of dynamical data in many different disciplines are all waiting to be mined with novel and reliable methods of n etwork reconstruction[1,2]. The most basic feature of a network is how the nodes are linked and how strong they are linked. This feature determines the overall behavior a nd affects the function of a network. However, this information of the wiring diagram of a network is often not available. Thus, a critical problem in the study of networks is how to infer the wiring diagram and reconstruct a network from measurements. In this talk, I will discuss our recent work on the challenging inverse problem of networ k reconstruction solely from the measurement of dynamical time-series data. We deriv ed general mathematical results relating the adjacency matrix of the network and the t ime-series measurements of the nodes. Based on these results, we have developed a m ethod that reconstructs both the links and their relative coupling strength using only t he time-series measurements of node dynamics as input[3]. We demonstrate that our method can give accurate results for unweighted and weighted random and scale-free networks with linear and nonlinear dynamics. We further show why relevance network s constructed using Pearson correlation coefficient and partial correlation coefficient ca n have significant deviations from the actual network. 1. "Extracting connectivity from dynamics for networks with uniform bidirectional coupling", E.S.C. Ching, Pik-Yin Lai, and C. Y. Leung, Phys. Rev. E 88, p.042817 (2013). 2. "Reconstruction of Network Structures from Repeating Patterns in Simulated Bursting Dynamics", H. Song, C.C. Chen, J.J. Sun, Pik-Yin Lai, and C. K. Chan, Phys. Rev. E 90, 012703 (2014). 3. "Reconstructing Weighted Networks from dynamics", E.S.C. Ching, Pik-Yin Lai, and C. Y. Leung, Phys. Rev. E (Rapid Comm.) 91, 030801 (2015).
  • [Talk 23] Phase equi...
    Julien Tailleur
    July 7(Thu), 2016
  • [Talk 22] Pressure i...
    Yariv Kafri
    July 7(Thu), 2016
  • [Talk 27] Height dis...
    Satya N. Majumdar
    July 7(Thu), 2016
  • [Talk 26] Noninterse...
    Dong Wang
    July 7(Thu), 2016
  • [Talk 24] Exact stat...
    Gregory Schehr
    July 7(Thu), 2016
  • [Talk 17] Adaptive w...
    Joachim Krug
    July 6(Wed), 2016
  • [Talk 20] Scaling co...
    S. H. Ebrahimnazhad Rahbari
    July 6(Wed), 2016
  • [Talk 18] Evolution ...
    Hyeong-Chai Jeong
    July 6(Wed), 2016
  • [Talk 15] Role of Bi...
    Meesoon Ha
    July 5(Tue), 2016
  • [Talk 16] Probabilit...
    Eunghyun Lee
    July 5(Tue), 2016
  • [Talk 14] Irreversib...
    Erick Lutz
    July 5(Tue), 2016
  • [Talk 12] Efficiency...
    Sang Hoon Lee
    July 5(Tue), 2016
  • [Talk 11] The Gibbs ...
    Yuto Murashita
    July 5(Tue), 2016
  • [Talk 9] Reconstruct...
    Pik-Yin Lai
    July 5(Tue), 2016
  • [Talk 8] Does the tr...
    Masaki Sano
    July 5(Tue), 2016
  • [Talk 1] Thermodynam...
    Shin-ichi Sasa
    July 4(Mon), 2016
  • [Talk 7] Universalit...
    Wonmuk Hwang
    July 4(Mon), 2016
  • [Talk 6] Nonequilibr...
    Leihan Tang
    July 4(Mon), 2016
  • [Talk 5] A non-equil...
    Julian Lee
    July 4(Mon), 2016
  • [Talk 4] Dynamical s...
    Yongjoo Baek
    July 4(Mon), 2016
  • [Talk 2] Fluctuation...
    Sanjib Sabhapandit
    July 4(Mon), 2016