Archives AI News

A Flexible Framework for Incorporating Patient Preferences Into Q-Learning

arXiv:2307.12022v2 Announce Type: replace Abstract: In real-world healthcare settings, treatment decisions often involve optimizing for multivariate outcomes such as treatment efficacy and severity of side effects based on individual preferences. However, existing statistical methods for estimating dynamic treatment regimes (DTRs) usually assume a univariate outcome, and the few methods that deal with composite outcomes suffer from limitations such as restrictions to a single time point and limited theoretical guarantees. To address these limitations, we propose Latent Utility Q-Learning (LUQ-Learning), a latent model approach that adapts Q-learning to tackle the aforementioned difficulties. Our framework allows for an arbitrary finite number of decision points and outcomes, incorporates personal preferences, and achieves asymptotic performance guarantees with realistic assumptions. We conduct simulation experiments based on an ongoing trial for low back pain as well as a well-known trial for schizophrenia. In both settings, LUQ-Learning achieves highly competitive performance compared to alternative baselines.

FBMS: An R Package for Flexible Bayesian Model Selection and Model Averaging

arXiv:2509.00753v1 Announce Type: cross Abstract: The FBMS R package facilitates Bayesian model selection and model averaging in complex regression settings by employing a variety of Monte Carlo model exploration methods. At its core, the package implements an efficient Mode Jumping Markov Chain Monte Carlo (MJMCMC) algorithm, designed to improve mixing in multi-modal posterior landscapes within Bayesian generalized linear models. In addition, it provides a genetically modified MJMCMC (GMJMCMC) algorithm that introduces nonlinear feature generation, thereby enabling the estimation of Bayesian generalized nonlinear models (BGNLMs). Within this framework, the algorithm maintains and updates populations of transformed features, computes their posterior probabilities, and evaluates the posteriors of models constructed from them. We demonstrate the effective use of FBMS for both inferential and predictive modeling in Gaussian regression, focusing on different instances of the BGNLM class of models. Furthermore, through a broad set of applications, we illustrate how the methodology can be extended to increasingly complex modeling scenarios, extending to other response distributions and mixed effect models.

Federated learning over physical channels: adaptive algorithms with near-optimal guarantees

arXiv:2509.02538v1 Announce Type: cross Abstract: In federated learning, communication cost can be significantly reduced by transmitting the information over the air through physical channels. In this paper, we propose a new class of adaptive federated stochastic gradient descent (SGD) algorithms that can be implemented over physical channels, taking into account both channel noise and hardware constraints. We establish theoretical guarantees for the proposed algorithms, demonstrating convergence rates that are adaptive to the stochastic gradient noise level. We also demonstrate the practical effectiveness of our algorithms through simulation studies with deep learning models.

Exam Readiness Index (ERI): A Theoretical Framework for a Composite, Explainable Index

arXiv:2509.00718v1 Announce Type: cross Abstract: We present a theoretical framework for an Exam Readiness Index (ERI): a composite, blueprint-aware score R in [0,100] that summarizes a learner's readiness for a high-stakes exam while remaining interpretable and actionable. The ERI aggregates six signals -- Mastery (M), Coverage (C), Retention (R), Pace (P), Volatility (V), and Endurance (E) -- each derived from a stream of practice and mock-test interactions. We formalize axioms for component maps and the composite, prove monotonicity, Lipschitz stability, and bounded drift under blueprint re-weighting, and show existence and uniqueness of the optimal linear composite under convex design constraints. We further characterize confidence bands via blueprint-weighted concentration and prove compatibility with prerequisite-admissible curricula (knowledge spaces / learning spaces). The paper focuses on theory; empirical study is left to future work.

Top-H Decoding: Adapting the Creativity and Coherence with Bounded Entropy in Text Generation

arXiv:2509.02510v1 Announce Type: cross Abstract: Large language models (LLMs), despite their impressive performance across a wide range of tasks, often struggle to balance two competing objectives in open-ended text generation: fostering diversity and creativity while preserving logical coherence. Existing truncated sampling techniques, including temperature scaling, top-$p$ (nucleus) sampling, and min-$p$ sampling, aim to manage this trade-off. However, they exhibit limitations, particularly in the effective incorporation of the confidence of the model into the corresponding sampling strategy. For example, min-$p$ sampling relies on a single top token as a heuristic for confidence, eventually underutilizing the information of the probability distribution. Toward effective incorporation of the confidence of the model, in this paper, we present **top-H** decoding. We first establish the theoretical foundation of the interplay between creativity and coherence in truncated sampling by formulating an **entropy-constrained minimum divergence** problem. We then prove this minimization problem to be equivalent to an **entropy-constrained mass maximization** (ECMM) problem, which is NP-hard. Finally, we present top-H decoding, a computationally efficient greedy algorithm to solve the ECMM problem. Extensive empirical evaluations demonstrate that top-H outperforms the state-of-the-art (SoTA) alternative of min-$p$ sampling by up to **25.63%** on creative writing benchmarks, while maintaining robustness on question-answering datasets such as GPQA, GSM8K, and MT-Bench. Additionally, an *LLM-as-judge* evaluation confirms that top-H indeed produces coherent outputs even at higher temperatures, where creativity is especially critical. In summary, top-H advances SoTA in open-ended text generation and can be *easily integrated* into creative writing applications. The code is available at https://github.com/ErfanBaghaei/Top-H-Decoding.

Context-Action Embedding Learning for Off-Policy Evaluation in Contextual Bandits

arXiv:2509.00648v1 Announce Type: cross Abstract: We consider off-policy evaluation (OPE) in contextual bandits with finite action space. Inverse Propensity Score (IPS) weighting is a widely used method for OPE due to its unbiased, but it suffers from significant variance when the action space is large or when some parts of the context-action space are underexplored. Recently introduced Marginalized IPS (MIPS) estimators mitigate this issue by leveraging action embeddings. However, these embeddings do not minimize the mean squared error (MSE) of the estimators and do not consider context information. To address these limitations, we introduce Context-Action Embedding Learning for MIPS, or CAEL-MIPS, which learns context-action embeddings from offline data to minimize the MSE of the MIPS estimator. Building on the theoretical analysis of bias and variance of MIPS, we present an MSE-minimizing objective for CAEL-MIPS. In the empirical studies on a synthetic dataset and a real-world dataset, we demonstrate that our estimator outperforms baselines in terms of MSE.

Calibration through the Lens of Indistinguishability

arXiv:2509.02279v1 Announce Type: cross Abstract: Calibration is a classical notion from the forecasting literature which aims to address the question: how should predicted probabilities be interpreted? In a world where we only get to observe (discrete) outcomes, how should we evaluate a predictor that hypothesizes (continuous) probabilities over possible outcomes? The study of calibration has seen a surge of recent interest, given the ubiquity of probabilistic predictions in machine learning. This survey describes recent work on the foundational questions of how to define and measure calibration error, and what these measures mean for downstream decision makers who wish to use the predictions to make decisions. A unifying viewpoint that emerges is that of calibration as a form of indistinguishability, between the world hypothesized by the predictor and the real world (governed by nature or the Bayes optimal predictor). In this view, various calibration measures quantify the extent to which the two worlds can be told apart by certain classes of distinguishers or statistical measures.

Feature Augmentations for High-Dimensional Learning

arXiv:2509.00232v1 Announce Type: cross Abstract: High-dimensional measurements are often correlated which motivates their approximation by factor models. This holds also true when features are engineered via low-dimensional interactions or kernel tricks. This often results in over parametrization and requires a fast dimensionality reduction. We propose a simple technique to enhance the performance of supervised learning algorithms by augmenting features with factors extracted from design matrices and their transformations. This is implemented by using the factors and idiosyncratic residuals which significantly weaken the correlations between input variables and hence increase the interpretability of learning algorithms and numerical stability. Extensive experiments on various algorithms and real-world data in diverse fields are carried out, among which we put special emphasis on the stock return prediction problem with Chinese financial news data due to the increasing interest in NLP problems in financial studies. We verify the capability of the proposed feature augmentation approach to boost overall prediction performance with the same algorithm. The approach bridges a gap in research that has been overlooked in previous studies, which focus either on collecting additional data or constructing more powerful algorithms, whereas our method lies in between these two directions using a simple PCA augmentation.

Probabilities of Causation and Root Cause Analysis with Quasi-Markovian Models

arXiv:2509.02535v1 Announce Type: new Abstract: Probabilities of causation provide principled ways to assess causal relationships but face computational challenges due to partial identifiability and latent confounding. This paper introduces both algorithmic simplifications, significantly reducing the computational complexity of calculating tighter bounds for these probabilities, and a novel methodological framework for Root Cause Analysis that systematically employs these causal metrics to rank entire causal paths.

Wild Refitting for Model-Free Excess Risk Evaluation of Opaque ML/AI Models under Bregman Loss

arXiv:2509.02476v1 Announce Type: new Abstract: We study the problem of evaluating the excess risk of classical penalized empirical risk minimization (ERM) with Bregman losses. We show that by leveraging the recently proposed wild refitting procedure (Wainwright, 2025), one can efficiently upper bound the excess risk through the so-called "wild optimism," without relying on the global structure of the underlying function class. This property makes our approach inherently model-free. Unlike conventional analyses, our framework operates with just one dataset and black-box access to the training procedure. The method involves randomized vector-valued symmetrization with an appropriate scaling of the prediction residues and constructing artificially modified outcomes, upon which we retrain a second predictor for excess risk estimation. We establish high-probability performance guarantees both under the fixed design setting and the random design setting, demonstrating that wild refitting under Bregman losses, with an appropriately chosen wild noise scale, yields a valid upper bound on the excess risk. This work thus is promising for theoretically evaluating modern opaque ML and AI models such as deep neural networks and large language models, where the model class is too complex for classical learning theory and empirical process techniques to apply.