pith. sign in
Pith Number

pith:4N7ABM2U

pith:2026:4N7ABM2UUZYZF3F36EJNP5OP3X
not attested not anchored not stored refs resolved

Robust Sequential Experimental Design for A/B Testing

Chengchun Shi, Hongtu Zhu, Niansheng Tang, Qianglin Wen, Ting Li, Xiangkun Wu, Yingying Zhang

Robust sequential experimental design bounds the worst-case mean squared error of estimated treatment effects in A/B testing under model misspecification.

arxiv:2605.12899 v1 · 2026-05-13 · stat.ML · cs.LG

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{4N7ABM2UUZYZF3F36EJNP5OP3X}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp
2 Internet Archive
3 Author claim open · sign in to claim
4 Citations open
5 Replications open
Portable graph bundle live · download bundle · merged state
The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

Theoretically, we prove that our design bounds the worst-case mean squared error of the estimated treatment effect.

C2weakest assumption

The unified framework assumes that model misspecification can be controlled within a single design that covers both contextual bandit and dynamic settings, but the abstract provides no details on the precise class of misspecification or the conditions required for the bound to hold.

C3one line summary

A unified robust framework for sequential A/B testing bounds the worst-case mean squared error of treatment effect estimates under model misspecification in both contextual bandit and dynamic regimes.

References

25 extracted · 25 resolved · 1 Pith anchors

[1] doi: 10.1093/jrsssb/qkad072. Atkinson, A. C. Optimum biased coin designs for sequential clinical trials with prognostic factors.Biometrika, 69(1): 61–67, · doi:10.1093/jrsssb/qkad072
[2] arXiv preprint arXiv:2112.13495 , year=
[3] URL https: //doi.org/10.1287/mnsc.2019.3424 2019 · doi:10.1287/mnsc.2019.3424
[4] Budgeted active experimen- tation for treatment effect estimation from observational and randomized data.arXiv preprint arXiv:2602.22021,
[5] Detecting interference in A/B testing with increasing allocation.arXiv preprint arXiv:2211.03262,
Receipt and verification
First computed 2026-05-18T03:09:10.766621Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

e37e00b354a67192ecbbf112d7f5cfddec09f483c9ddfb6999752e88d6ca355f

Aliases

arxiv: 2605.12899 · arxiv_version: 2605.12899v1 · doi: 10.48550/arxiv.2605.12899 · pith_short_12: 4N7ABM2UUZYZ · pith_short_16: 4N7ABM2UUZYZF3F3 · pith_short_8: 4N7ABM2U
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/4N7ABM2UUZYZF3F36EJNP5OP3X \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: e37e00b354a67192ecbbf112d7f5cfddec09f483c9ddfb6999752e88d6ca355f
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "ac7f3abda1c207275f824ce485f77e876d5ba85470924061ab5c3b6f26205530",
    "cross_cats_sorted": [
      "cs.LG"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "stat.ML",
    "submitted_at": "2026-05-13T02:24:57Z",
    "title_canon_sha256": "7346a51bd392f998a0063714cadaeae5b23ff6c39a3e59ea82cfeb86b831ca80"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.12899",
    "kind": "arxiv",
    "version": 1
  }
}