Decision-Making under Uncertainty

Overview

Decision-making under uncertainty studies how agents should choose actions when outcomes, states, preferences, or even the governing model are imperfectly known. Common approaches include Bayesian beliefs and partially observable Markov decision processes for sequential learning, robust optimization for misspecified distributions, risk-sensitive criteria such as CVaR, and conformal methods that calibrate when to act, gather information, or abstain. These methods are tested in settings such as autonomous driving and robotics, clinical diagnosis, energy and supply-chain scheduling, finance, and human–AI interaction. A central tension is that better-calibrated or more conservative uncertainty estimates do not automatically improve decisions: methods must balance safety and tail risk against utility, computational cost, and the value of additional information.

Latest papers

213 papers13 papers in the last four weeks, down 38% on the four weeks before. 0.1% of all new papers.

May 29, 2024

Active Exploration via Autoregressive Generation of Missing Data

We pose uncertainty quantification and exploration in online decision-making as a problem of training and generation from an autoregressive sequence model, an area experiencing rapid innovation. Our approach rests on viewing uncertainty as arising from missing future outcomes that could be revealed through action choices, rather than from unobservable latent parameters of the environment. This reformulation aligns naturally with modern machine learning capabilities: we can i) train generative models through next-outcome prediction rather than fit explicit priors, ii) assess uncertainty through autoregressive generation rather than sampling latent parameters from posteriors, and iii) adapt to new information by extending the sequence model's context rather than explicit posterior updating. Our main theoretical result establishes a reduction from online decision-making to offline next-outcome prediction: Bayesian regret is controlled directly by the sequence model's offline prediction loss, without requiring an explicit latent-variable posterior. Experiments, including a semi-synthetic news recommendation task, show that autoregressive generation produces calibrated epistemic uncertainty and enables effective exploration by using article text as prior information to focus exploration on resolving remaining uncertainties.
Nov 30, 2021

Double Fuzzy Probabilistic Interval Linguistic Term Set and a Dynamic Fuzzy Decision Making Model based on Markov Process with tts Application in Multiple Criteria Group Decision Making

The probabilistic linguistic term has been proposed to deal with probability distributions in provided linguistic evaluations. However, because it has some fundamental defects, it is often difficult for decision-makers to get reasonable information of linguistic evaluations for group decision making. In addition, weight information plays a significant role in dynamic information fusion and decision making process. However, there are few research methods to determine the dynamic attribute weight with time. In this paper, I propose the concept of double fuzzy probability interval linguistic term set (DFPILTS). Firstly, fuzzy semantic integration, DFPILTS definition, its preference relationship, some basic algorithms and aggregation operators are defined. Then, a fuzzy linguistic Markov matrix with its network is developed. Then, a weight determination method based on distance measure and information entropy to reducing the inconsistency of DFPILPR and obtain collective priority vector based on group consensus is developed. Finally, an aggregation-based approach is developed, and an optimal investment case from a financial risk is used to illustrate the application of DFPILTS and decision method in multi-criteria decision making.
Sep 24, 2020

Learning in a Small/Big World

Complexity and limited ability affect how we learn and make decisions under uncertainty. Using finite automata to model belief formation, this paper studies the characteristics of optimal learning behavior in small and big worlds, where the complexity of the environment is low and high, respectively, relative to the cognitive ability of the decision maker. The optimal long-run average payoff and distribution of actions are well approximated by the Bayesian benchmark in very small worlds, but the gap widens as the worlds get bigger. In big worlds, it can be optimal for the decision maker to ignore some states entirely, concentrating limited cognitive resources on a subset of states. This ``optimal ignorance'' is reminiscent of well-documented non-Bayesian learning patterns such as the use of heuristics, correlation neglect, and inattentive learning. The results establish a relationship between complexity, cognitive ability, and the scope for non-Bayesian learning behavior.