The paper introduces the Descriptive‑Complexity Information Criterion (DCIC), a new framework for selecting models when predictors are highly correlated and the model class is uncertain. DCIC uses Kraft‑admissible code lengths to regularize large collections of candidate models, achieving selection consistency under sub‑Weibull noise without requiring RIP‑type conditions and providing non‑asymptotic oracle risk bounds even when the model is misspecified. The approach also unifies heterogeneous model classes on a common complexity scale, enables class–model recovery under identifiability conditions, and offers a complexity‑guided search path that balances computational effort with statistical accuracy, as demonstrated by numerical experiments.
arXiv:2609.01355v1 Announce Type: cross Abstract: Scenario optimization, conformal prediction, and related distribution-free certification methods use finite samples to construct decisions or predict...
arXiv:2303. 08777v3 Announce Type: replace-cross Abstract: Cross-validation is one of the most widely used tools for risk estimation and model selection in statistics and machine learning, yet its theoretical properties when embedded in a learning procedure remain insufficiently understood.
arXiv:2607. 23454v1 Announce Type: new Abstract: Data-driven remaining useful life (RUL) prediction requires complete degradation trajectories for training, yet such run-to-failure data are scarce and expensive.
arXiv:2606. 14690v1 Announce Type: new Abstract: We study a \emph{max-risk} objective for active learning in a multi-group mean estimation $d$-armed bandits: a learner adaptively allocates a budget of $T$ samples across $d$ groups to minimize the worst-case uncertainty index $\max_{k\in[d]}\sigma_k^2/n_k$, where $\sigma_k$ is the standard deviation of the distribution of arm $d$, and $n_k$ is the number of times arm $d$ is sampled.
arXiv:2608. 13514v1 Announce Type: cross Abstract: We revisit the problem of learning predictors robust to adversarial examples at test-time.
arXiv:2606. 17319v1 Announce Type: cross Abstract: Motivated by the optimization of bounded binary black-box functions, we study the problem of learning polynomial surrogates over the Boolean hypercube.
arXiv:2506. 11336v2 Announce Type: replace Abstract: We study the sample complexity of stochastic convex optimization when problem parameters such as the distance to optimality and the Lipschitz constant are unknown.
arXiv:2609.23937v1 Announce Type: cross Abstract: Robust linear fits can resist response contamination yet remain too dense or unstable for useful global explanations. We propose penalized distillati...
We revisit the problem of learning predictors robust to adversarial examples at test-time. We prove that VC classes are adversarially robustly learnable with sample complexity linear in the VC dimension $d$, providing an exponential improvement over the previous upper bound of Montasser, Hanneke, and Srebro (2019).
arXiv:2607. 05791v1 Announce Type: cross Abstract: Boosting is a fundamental technique for generically improving the accuracy of learning algorithms (Schapire 1989).
arXiv:2606. 20557v1 Announce Type: new Abstract: A model is multicalibrated on a collection of group weights $G$ if it is calibrated -- i.
The paper introduces Resolution-Aware Experimental Design (RAED), a method that selects experiments by minimizing the expected size of the nonempty structural candidate set while controlling false-exclusion rates. RAED is shown to preserve expected ordering under a composite Blackwell comparison and is implemented via a learned score-based approach with finite-sample nuisance-average and positive-tail calibration. Experiments on subsurface-flow, fluvial, and methane-oxidation benchmarks demonstrate RAED’s ability to resolve structural ambiguities and provide finite-sample guarantees for tail-sensitive nuisance risk.