Skip to content

Update paper to current Adaptive and fix errors #1

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
basnijholt wants to merge 4 commits into
base: master
Choose a base branch
from
Open

Update paper to current Adaptive and fix errors #1

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
e6444aa
Update paper to current Adaptive and fix errors
basnijholt Jun 10, 2026
a758df6
Name the popped loss max_loss instead of _
basnijholt Jun 10, 2026
dc710f9
Bump adaptive_repo to current main and soften unbenchmarked claims
basnijholt Jun 10, 2026
4890ef4
Quantify isoline accuracy with mean/Hausdorff distance
basnijholt Jun 10, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion adaptive_repo
Open in desktop
Submodule adaptive_repo updated 140 files
22 changes: 21 additions & 1 deletion figures.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -947,9 +947,29 @@
" # Isolines\n",
" vertices, lines = learner._get_iso(level, which=\"line\")\n",
" paths = np.array([[vertices[i], vertices[j]] for i, j in lines]).T\n",
" print(\"{} line segments\".format(len(paths.T)))\n",
" ax.plot(*paths, c=\"k\", lw=0.7)\n",
"\n",
" # Distance between the estimated isoline and the true isoline,\n",
" # y = cbrt(level - x^2), resampling both densely\n",
" from scipy.spatial import cKDTree\n",
"\n",
" xs_true = np.linspace(*bounds[0], 200_001)\n",
" true_pts = np.column_stack([xs_true, np.cbrt(level - xs_true ** 2)])\n",
" seg_pts = []\n",
" for i_, j_ in lines:\n",
" p, q = np.asarray(vertices[i_]), np.asarray(vertices[j_])\n",
" n = max(2, int(np.ceil(np.linalg.norm(q - p) / 1e-3)))\n",
" seg_pts.append(p + np.linspace(0, 1, n)[:, None] * (q - p))\n",
" seg_pts = np.vstack(seg_pts)\n",
" d_est = cKDTree(true_pts).query(seg_pts)[0]\n",
" d_true = cKDTree(seg_pts).query(true_pts)[0]\n",
" mean_dist = (d_est.mean() + d_true.mean()) / 2\n",
" hausdorff = max(d_est.max(), d_true.max())\n",
" print(\n",
" f\"{kind}: {len(lines)} segments, mean distance {mean_dist:.2g}, \"\n",
" f\"Hausdorff distance {hausdorff:.2g}\"\n",
" )\n",
"\n",
" values = np.array(list(learner.data.values()))\n",
" ax.imshow(\n",
" learner.plot(npoints if kind == \"homogeneous\" else None).Image.I.data,\n",
Expand Down
2 changes: 2 additions & 0 deletions pandoc/revtex.template
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -84,7 +84,9 @@
$for(author)$
$if(author.name)$
\author{$author.name$}
$if(author.email)$
\email[Electronic address: ]{$author.email$}
$endif$
$for(author.affiliation)$
\affiliation{$author.affiliation$}
$endfor$
Expand Down
26 changes: 26 additions & 0 deletions paper.bib
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -141,6 +141,19 @@ @article{Bommer2019
journal = {Phys. Rev. Lett.}
}

@article{Chaloner1995,
doi = {10.1214/ss/1177009939},
year = 1995,
month = {aug},
publisher = {Institute of Mathematical Statistics},
volume = {10},
number = {3},
pages = {273--304},
author = {Kathryn Chaloner and Isabella Verdinelli},
title = {Bayesian Experimental Design: A Review},
journal = {Stat. Sci.}
}

@article{Chen2017,
doi = {10.1088/1361-6501/aa7d31},
year = 2017,
Expand Down Expand Up @@ -281,6 +294,19 @@ @article{Nijholt2016
journal = {Phys. Rev. B}
}

@article{Ryan2015,
doi = {10.1111/insr.12107},
year = 2015,
month = {jun},
publisher = {Wiley},
volume = {84},
number = {1},
pages = {128--154},
author = {Elizabeth G. Ryan and Christopher C. Drovandi and James M. McGree and Anthony N. Pettitt},
title = {A Review of Modern Computational Algorithms for Bayesian Optimal Design},
journal = {Int. Stat. Rev.}
}

@article{Visvalingam1990,
doi = {10.1111/j.1467-8659.1990.tb00398.x},
year = 1990,
Expand Down
Loading