Registered Data

[00694] Large deviation theory-based adaptive importance sampling for rare events in high dimensions

  • Session Time & Room : 5B (Aug.25, 10:40-12:20) @E505
  • Type : Contributed Talk
  • Abstract : I will discuss our proposed method for estimating rare event probabilities for expensive-to-evaluate numerical models in high dimensions. The approach combines ideas from large deviation theory and adaptive importance sampling. Large deviation theory is used to find a good initial biasing distribution and to identify a low-dimensional subspace that is most informative of the rare event probability. We compare the method with a state-of-the-art cross-entropy-based importance sampling scheme.
  • Classification : 65C05, 60F10, 62L12, 65F15, 65K10
  • Format : Talk at Waseda University
  • Author(s) :
    • Shanyin Tong (Columbia University)
    • Georg Stadler (New York University)