Balanced and Robust Multi-Treatment Experimental Designs via Randomized Differencing

Qing Chen, Jing Jia, Peng Zhang
Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, PMLR 300:1468-1476, 2026.

Abstract

We introduce GKK+, a new design for multi-arm randomized controlled trials. Standard Bernoulli randomization is robust but often yields poor covariate balance, while existing restricted-randomness designs mainly address two-arm settings. GKK+ extends the Karmarkar–Karp (KK) differencing method to multiple arms. When covariates are smooth and well-behaved, GKK+ achieves an exponentially better covariate balance than the standard Bernoulli design while preserving sufficient randomness. GKK+ improves efficiency in estimating treatment effects and supports standard asymptotic inference. Simulations on synthetic and real datasets demonstrate improved balance and lower estimator variance compared to existing methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v300-chen26b, title = { Balanced and Robust Multi-Treatment Experimental Designs via Randomized Differencing }, author = {Chen, Qing and Jia, Jing and Zhang, Peng}, booktitle = {Proceedings of The 29th International Conference on Artificial Intelligence and Statistics}, pages = {1468--1476}, year = {2026}, editor = {Khan, Emtiyaz and Li, Yingzhen and Solin, Arno and Ramdas, Aaditya}, volume = {300}, series = {Proceedings of Machine Learning Research}, month = {02--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v300/main/assets/chen26b/chen26b.pdf}, url = {https://proceedings.mlr.press/v300/chen26b.html}, abstract = { We introduce GKK+, a new design for multi-arm randomized controlled trials. Standard Bernoulli randomization is robust but often yields poor covariate balance, while existing restricted-randomness designs mainly address two-arm settings. GKK+ extends the Karmarkar–Karp (KK) differencing method to multiple arms. When covariates are smooth and well-behaved, GKK+ achieves an exponentially better covariate balance than the standard Bernoulli design while preserving sufficient randomness. GKK+ improves efficiency in estimating treatment effects and supports standard asymptotic inference. Simulations on synthetic and real datasets demonstrate improved balance and lower estimator variance compared to existing methods. } }
Endnote
%0 Conference Paper %T Balanced and Robust Multi-Treatment Experimental Designs via Randomized Differencing %A Qing Chen %A Jing Jia %A Peng Zhang %B Proceedings of The 29th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2026 %E Emtiyaz Khan %E Yingzhen Li %E Arno Solin %E Aaditya Ramdas %F pmlr-v300-chen26b %I PMLR %P 1468--1476 %U https://proceedings.mlr.press/v300/chen26b.html %V 300 %X We introduce GKK+, a new design for multi-arm randomized controlled trials. Standard Bernoulli randomization is robust but often yields poor covariate balance, while existing restricted-randomness designs mainly address two-arm settings. GKK+ extends the Karmarkar–Karp (KK) differencing method to multiple arms. When covariates are smooth and well-behaved, GKK+ achieves an exponentially better covariate balance than the standard Bernoulli design while preserving sufficient randomness. GKK+ improves efficiency in estimating treatment effects and supports standard asymptotic inference. Simulations on synthetic and real datasets demonstrate improved balance and lower estimator variance compared to existing methods.
APA
Chen, Q., Jia, J. & Zhang, P.. (2026). Balanced and Robust Multi-Treatment Experimental Designs via Randomized Differencing . Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 300:1468-1476 Available from https://proceedings.mlr.press/v300/chen26b.html.

Related Material