What Capable Agents Must Know: Selection Theorems for Robust Decision-Making under Uncertainty

• URL: http://arxiv.org/abs/2603.02491v1
• Date: Tue, 03 Mar 2026 00:47:58 GMT
• Title: What Capable Agents Must Know: Selection Theorems for Robust Decision-Making under Uncertainty
• Authors: Aran Nayebi,
• Abstract summary: We prove "selection theorems" showing that low "average-case regret" forces an agent to implement a predictive, structured internal state.<n>We show that regret bounds limit probability mass on suboptimal bets, enforcing the predictive distinctions needed to separate high-margin outcomes.
• Score: 1.6868147729303773
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: As artificial agents become increasingly capable, what internal structure is *necessary* for an agent to act competently under uncertainty? Classical results show that optimal control can be *implemented* using belief states or world models, but not that such representations are required. We prove quantitative "selection theorems" showing that low *average-case regret* on structured families of action-conditioned prediction tasks forces an agent to implement a predictive, structured internal state. Our results cover stochastic policies, partial observability, and evaluation under task distributions, without assuming optimality, determinism, or access to an explicit model. Technically, we reduce predictive modeling to binary "betting" decisions and show that regret bounds limit probability mass on suboptimal bets, enforcing the predictive distinctions needed to separate high-margin outcomes. In fully observed settings, this yields approximate recovery of the interventional transition kernel; under partial observability, it implies necessity of belief-like memory and predictive state, addressing an open question in prior world-model recovery work.

Related papers

• The Illusion of Certainty: Uncertainty quantification for LLMs fails under ambiguity [48.899855816199484]
  We introduce MAQA* and AmbigQA*, the first ambiguous question-answering (QA) datasets equipped with ground-truth answer distributions.<n>We show that predictive-distribution and ensemble-based estimators are fundamentally limited under ambiguity.
  arXiv  Detail & Related papers  (2025-11-06T14:46:35Z)
• How Far Are We from Generating Missing Modalities with Foundation Models? [49.425856207329524]
  We propose an agentic framework tailored for missing modality reconstruction.<n>Our method reduces FID for missing image reconstruction by at least 14% and MER for missing text reconstruction by at least 10% compared to baselines.
  arXiv  Detail & Related papers  (2025-06-04T03:22:44Z)
• Conformalized Decision Risk Assessment [5.391713612899277]
  We introduce CREDO, a novel framework that quantifies for any candidate decision, a distribution-free upper bound on the probability that the decision is suboptimal.<n>By combining inverse optimization geometry with conformal prediction and generative modeling, CREDO produces risk certificates that are both statistically rigorous and practically interpretable.
  arXiv  Detail & Related papers  (2025-05-19T15:24:38Z)
• Model uncertainty quantification using feature confidence sets for outcome excursions [0.0]
  This paper introduces a novel, model-agnostic framework for quantifying uncertainty in continuous and binary outcomes.<n>It is validated through simulations and applied to real-world datasets in contexts such as housing price prediction and time to sepsis diagnosis in healthcare.
  arXiv  Detail & Related papers  (2025-04-28T04:08:07Z)
• Truthful Elicitation of Imprecise Forecasts [11.153198087930756]
  We propose a framework for scoring imprecise forecasts -- forecasts given as a set of beliefs.<n>We show that truthful elicitation of imprecise forecasts is achievable using proper scoring rules randomized over the aggregation procedure.
  arXiv  Detail & Related papers  (2025-03-20T17:53:35Z)
• Calibrating Neural Simulation-Based Inference with Differentiable Coverage Probability [50.44439018155837]
  We propose to include a calibration term directly into the training objective of the neural model. By introducing a relaxation of the classical formulation of calibration error we enable end-to-end backpropagation. It is directly applicable to existing computational pipelines allowing reliable black-box posterior inference.
  arXiv  Detail & Related papers  (2023-10-20T10:20:45Z)
• Discretization-Induced Dirichlet Posterior for Robust Uncertainty Quantification on Regression [17.49026509916207]
  Uncertainty quantification is critical for deploying deep neural networks (DNNs) in real-world applications. For vision regression tasks, current AuxUE designs are mainly adopted for aleatoric uncertainty estimates. We propose a generalized AuxUE scheme for more robust uncertainty quantification on regression tasks.
  arXiv  Detail & Related papers  (2023-08-17T15:54:11Z)
• When Does Confidence-Based Cascade Deferral Suffice? [69.28314307469381]
  Cascades are a classical strategy to enable inference cost to vary adaptively across samples. A deferral rule determines whether to invoke the next classifier in the sequence, or to terminate prediction. Despite being oblivious to the structure of the cascade, confidence-based deferral often works remarkably well in practice.
  arXiv  Detail & Related papers  (2023-07-06T04:13:57Z)
• Uncertainty estimation under model misspecification in neural network regression [3.2622301272834524]
  We study the effect of the model choice on uncertainty estimation. We highlight that under model misspecification, aleatoric uncertainty is not properly captured.
  arXiv  Detail & Related papers  (2021-11-23T10:18:41Z)
• Dense Uncertainty Estimation [62.23555922631451]
  In this paper, we investigate neural networks and uncertainty estimation techniques to achieve both accurate deterministic prediction and reliable uncertainty estimation. We work on two types of uncertainty estimations solutions, namely ensemble based methods and generative model based methods, and explain their pros and cons while using them in fully/semi/weakly-supervised framework.
  arXiv  Detail & Related papers  (2021-10-13T01:23:48Z)
• Alternative Microfoundations for Strategic Classification [33.67797984699066]
  We show that rational agents with perfect information produce discontinuities in the aggregate response to a decision rule. optimal decision rules under standard microfoundations maximize a measure of negative externality known as social burden. Our model retains analytical tractability, leads to more robust insights about stable points, and imposes a lower social burden at optimality.
  arXiv  Detail & Related papers  (2021-06-24T00:30:58Z)
• Heterogeneous-Agent Trajectory Forecasting Incorporating Class Uncertainty [54.88405167739227]
  We present HAICU, a method for heterogeneous-agent trajectory forecasting that explicitly incorporates agents' class probabilities. We additionally present PUP, a new challenging real-world autonomous driving dataset. We demonstrate that incorporating class probabilities in trajectory forecasting significantly improves performance in the face of uncertainty.
  arXiv  Detail & Related papers  (2021-04-26T10:28:34Z)
• Estimating Generalization under Distribution Shifts via Domain-Invariant Representations [75.74928159249225]
  We use a set of domain-invariant predictors as a proxy for the unknown, true target labels. The error of the resulting risk estimate depends on the target risk of the proxy model.
  arXiv  Detail & Related papers  (2020-07-06T17:21:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.