Computation

Haixiang Sun, Andrew Liu

This work addresses the problem of generating realistic and decision-relevant scenarios for stochastic programming, a key challenge in operations research and decision-making under uncertainty.

Steve Hanneke, Anay Mehrotra, Grigoris Velegkas et al.

Provides a theoretical characterization and first algorithm for a classic but understudied learning model, clarifying the role of membership queries.

Lifu Wei, Yinuo Ren, Naichen Shi et al.

It provides a computationally efficient and unbiased method for inference-time guidance in diffusion models, addressing the bottleneck of repeated score/gradient evaluations.

Buu Phan, Gergely Flamich, Ashish Khisti et al.

For practitioners using MCMC, this provides a more efficient coupling-based convergence diagnostic that scales better to high dimensions.

Kyurae Kim, Samuel Gruffaz, Ji Won Park et al.

For researchers using Langevin Monte Carlo for sampling, this work extends theoretical guarantees to the overdamped regime, showing the exponential integrator remains stable and effective.

Wenqian Xu, Feng Ji

For psychometricians and educational test developers, the RLMM enables scalable measurement of latent traits from interactive assessment process data, addressing a key bottleneck in applying MDP-based models to realistic tasks.

Qianyu Julie Zhu, Youssef Marzouk, Konstantinos Spiliopoulos et al.

For practitioners of Markov chain Monte Carlo, this work provides a principled way to design irreversible perturbations that accelerate convergence while controlling bias, addressing a gap in non-Gaussian and discretized settings.

Jungang Zou, Alex Ziyu Jiang, Qixuan Chen

This work addresses the coding bottleneck in MCMC workflows for Bayesian practitioners, but the results are preliminary and the system's capability is limited to built-in blocks.

Daniel Sharp, Bart van Bloemen Waanders, Youssef Marzouk

This addresses computational challenges in Bayesian inference for researchers in fields like inverse problems, but it is incremental as it builds on existing transport and annealing methods.

Fabian Zaiser, Jack Czenszak, Martin C. Rinard et al.

For developers of probabilistic programming systems, this work provides a principled, modular approach to incremental inference that improves scalability and correctness over ad-hoc methods.

Lucas H. McCabe, H. Howie Huang

Provides a drop-in replacement for entropy estimation in small-sample scenarios, benefiting fields like ecology and AI where data is scarce.

William Rudman, Abhishek Divekar, Kanishk Jain et al. · amazon-science

For scientists and researchers needing automated statistical modeling, VESTA addresses the bottleneck of model refinement by enabling active data exploration and tool creation, though the gains are incremental over existing agent-based systems.

Di Wu, Ling Liang, Haizhao Yang

For practitioners in resource-constrained experimental design, this work provides a more robust and scalable BOED approach that overcomes fundamental limitations of KL-based methods.

Aditya Dendukuri, Shivkumar Chandrasekaran, Linda Petzold

It addresses the challenge of efficiently solving the Chemical Master Equation for multiscale systems with widely separated reaction rates, which is important for computational systems biology.

Valerie N. P. Ho, Art B. Owen

Provides a highly effective variance reduction method for Monte Carlo simulation of Dirichlet boundary value problems, with substantial practical gains.

Nobuki Takayama, Takaharu Yaguchi, Yi Zhang

This work addresses the computational challenge of evaluating normalizing constants for statisticians, but it is incremental as it compares existing methods rather than introducing new ones.

Daniel Paulin, Peter A. Whalley

For researchers in MCMC and sampling methods, this work offers tighter theoretical guarantees for stochastic gradient sampling, addressing a previously open problem.

Hany Abdulsamad, Sahel Iqbal, Christian A. Naesseth et al.

This work addresses the problem of robust experimental design for researchers in Bayesian statistics and machine learning, offering a novel method to handle model misspecification, though it is incremental in building on existing information-theoretic concepts.

Dario Draca, Takuo Matsubara, Minh-Ngoc Tran

This work addresses optimization challenges for machine learning models with constrained parameters, such as those in variational inference, by enabling efficient natural gradient descent on manifolds.

Xun Huan, Jayanth Jagalur, Youssef Marzouk

For researchers and practitioners in modeling and prediction across sciences and engineering, this survey provides a comprehensive overview of OED methods and identifies key open problems.