protoplaid

3 ms·

1. 1.▲
   by protoplaid 6y ago

   > especially if the first sample looks particularly “good”. You've precisely described the problem: the algorithm will get stuck on a point if the first sample looks good and the assumption of zero variance. Until it randomly hits
2. 2.▲
   by protoplaid 6y ago

   line 62: exp_imp[sigma == 0.0] = 0.0 I'm afraid it never samples points more than once, since it estimated already-sampled-points as points with variance zero, and no expected improvement. IMHO that's wrong. Variance of a single
3. 3.▲
   by protoplaid 6y ago

   Correct me if I'm wrong, but it seems the bayesian_optimization.py optimizer in this library assumes that the sampled points are exact, ie their variance is zero. It doesn't seem to re-sample existing points. This will cause the a
4. 4.▲
   by protoplaid 6y ago

   Which algorithm would you recommend when the objective function is noisy (and nondeterministic)? For example the objective function is the "score" of a particular stochastic simulation, which can be started with varied initial ran