From d99fc7e60a6a9fb2c21a01bfa6dd998cdbcf8c81 Mon Sep 17 00:00:00 2001 From: Bas Nijholt <basnijholt@gmail.com> Date: Wed, 18 Sep 2019 13:25:14 +0200 Subject: [PATCH] anisotropic meshing --- paper.md | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/paper.md b/paper.md index 747b186..3e747be 100755 --- a/paper.md +++ b/paper.md @@ -325,7 +325,11 @@ For more details on how to use Adaptive, we recommend reading the tutorial insid # Possible extensions #### Anisotropic triangulation would improve the algorithm. -[@dyn1990data] +The current implementation of choosing the candidate point inside a simplex (triangle for 2D) with the highest loss, for the `LearnerND`, works by either picking a point (1) in the center of the simplex or (2) by picking a point on the longest edge of the simplex. +The choice depends on the shape of the simplex, where the algorithm tries to create regular simplices. +Alternatively, a good strategy is choosing points somewhere on the edge of a triangle such that the simplex aligns with the gradient of the function; creating an anisotropic triangulation[@dyn1990data]. +This is a similar approach to the anisotropic meshing techniques mentioned in the literature review. +We have started to implement this, however, there are still some unsolved problems. #### Learning stochastic functions is a promising direction. -- GitLab