New method for finding optimal treatment regimes in medical settings with time-varying unobserved factors.
problem Finding optimal treatment regimes in medical settings with time-varying unobserved factors.
method Extend Dynamic Treatment Regimes (DTRs) to Ambiguous Dynamic Treatment Regimes (ADTRs), connect to Ambiguous Partially Observable Mark Decision Processes (APOMDPs), and develop Reinforcement Learning methods.
result Established theoretical results for learning methods, including consistency and asymptotic normality.
Unified framework for DRO and DTA using Bayesian nonparametrics.
problem Combining DRO and DTA under ambiguity.
method Unified framework using DP and HDPs, with outlier robustness.
result Favorable performance in prediction accuracy and stability.
We characterize value functions in partially observable MDPs as semi-algebraic sets.
problem Understanding feasible value functions in partially observable Markov decision processes.
method Characterization of feasible value functions as semi-algebraic sets defined by polynomial inequalities.
result The feasible set of value functions in POMDPs is a semi-algebraic set, not a polytope as in MDPs.
New method optimises worst-case risk under model uncertainty.
problem Minimizing expected risk under posterior beliefs leads to sub-optimal decisions due to model uncertainty.
method Distributionally Robust Optimisation with Bayesian Ambiguity Sets (DRO-BAS)
result Improved out-of-sample robustness in the Newsvendor problem.
New algorithms solve robust MDPs efficiently, significantly faster than existing methods.
problem Computing robust MDP solutions with uncertainty in transition probabilities is computationally expensive.
method Partial policy iteration and fast robust Bellman operator computation methods.
result The proposed methods are many orders of magnitude faster than state-of-the-art approaches.
We solve a broad class of sequential decision-making problems with partially observed states.
problem Sequential decision-making under uncertainty with partially observed states.
method Modeling as a partially observed Markov decision process (POMDP) and separating state and modulation process.
result The approach allows for specialized approximate solution procedures.
The paper develops methods to estimate POMDPs from partial information.
problem Making decisions under partial information about state variables.
method Structural estimation of POMDP primitives using observable history.
result Conditions for model identifiability without state dynamics knowledge.
New algorithm improves reinforcement learning from partial observations.
problem Inferior performance of algorithms in real-world reinforcement learning due to partial observability.
method Representation-based approach to POMDPs, leading to a tractable algorithm.
result Empirically demonstrates superior performance with partial observations.
In this paper we consider stochastic optimization problems for an ambiguity averse decision maker who is uncertain about the parameters of the underlying process. In a first part we consider problems of optimal stopping under drift ambiguity for one-dimensional diffusion processes. Analogously to the case of ordinary o…
Improves DRO with Bayesian Ambiguity Sets for model misspecification.
problem Overly conservative decisions due to misspecified models in DRO.
method Introduces DRO-RoBAS with robust posterior predictive distribution.
result Outperforms other Bayesian and empirical DRO approaches in out-of-sample performance.
A geometric account explains why 'The Dress' is ambiguous, predicting observable signatures in image processing.
problem Understanding and predicting ambiguity in image processing, particularly in intrinsic image decomposition.
method Geometric analysis of intrinsic image decomposition, focusing on the discontinuous switch in prior-mode sections.
result Predicted signatures in albedo Jacobian and Fernet curvature can be observed in various models and datasets.
The Machina thought experiments pose to major non-expected utility models challenges that are similar to those posed by the Ellsberg thought experiments to subjective expected utility theory (SEUT). We test human choices in the `Ellsberg three-color example', confirming typical ambiguity aversion patterns, and the `Mac…
New algorithm for partially observable contexts in finance.
problem Decision making based on partially observable, correlated market information.
method EMKF-Bandit algorithm integrating system identification, filtering, and bandit algorithms.
result Sub-linear regret under conditions on filtering.
We investigate the optimal reinsurance problem under the criterion of maximizing the expected utility of terminal wealth when the insurance company has restricted information on the loss process. We propose a risk model with claim arrival intensity and claim sizes distribution affected by an unobservable environmental …
A new algorithm trains experts to safely guide agents in partially observed environments.
problem Existing imitation learning methods for POMDPs can lead to sub-optimal or unsafe policies.
method Derive an objective to encourage the expert to maximize the agent's reward, then use it to train both expert and agent.
result The algorithm produces an expert policy that the agent can safely imitate, outperforming fixed expert policies.
RLHF fails when humans only partially observe, leading to inflated or overjustified feedback.
problem Failure of reinforcement learning from human feedback in partially observable environments.
method Formal definition of failure cases, modeling human as Boltzmann rational, analyzing information provided by feedback.
result RLHF can deceptively inflate or overjustify feedback when humans have partial observations.
Linear recurrent networks explain reinforcement learning performance in partially observable settings.
problem Understanding why linear recurrent networks work in reinforcement learning with partial observability.
method Constructed and studied two linear filters for HMMs and action-controlled HMMs.
result Linear filters serve as sufficient statistics and reduce state ambiguity, explaining empirical reinforcement learning success.
This paper analyzes risk-sensitive reinforcement learning with Conditional Value-at-Risk (CVaR) for robust Markov Decision Processes.
problem Risk-sensitive reinforcement learning for robust Markov Decision Processes (RMDPs) with state-action-dependent ambiguity sets.
method The paper establishes a connection between robustness and risk sensitivity, defining a new risk measure NCVaR and proposing value iteration algorithms.
result The proposed approach using NCVaR optimization and value iteration algorithms can solve problems with state-action-dependent ambiguity sets.
New method identifies flawed internal models of the world in animals.
problem How animals make decisions with partial sensory information.
method Generalizes Inverse Rational Control to continuous nonlinear dynamics and noise.
result Identifies the best internal model explaining an agent's actions.
A new metric detects non-Markovian states in partially observable environments.
problem Learning state representations in partially observable environments.
method Introducing the λ-discrepancy metric to detect non-Markovian states. result The λ-discrepancy is zero for Markov processes and non-zero for partially observable environments. Methodology for estimating marked Hawkes processes with neural networks.
problem Estimating conditional intensity of marked Hawkes processes.
method Proposes two models: Shallow Neural Hawkes with marks and Neural Network for Non-Linear Hawkes with Marks.
result Validation on synthetic datasets and real-world cryptocurrency order book data.
A reject option improves partial-label learning's accuracy.
problem Ambiguously labeled data in real-world applications.
method Risk-consistent nearest-neighbor algorithm with a reject option.
result Our method provides the best trade-off between non-rejected predictions' number and accuracy.
Paper solves POMDPs in continuous time and discrete spaces.
problem Optimal decision making in discrete state and action space systems under partial observability.
method Combining optimal filtering theory and deep learning to solve a Hamilton-Jacobi-Bellman equation.
result Derives a mathematical description and solution approach for continuous-time POMDPs.
Partially observable Markov decision processes (POMDPs) with continuous state and observation spaces have powerful flexibility for representing real-world decision and control problems but are notoriously difficult to solve. Recent online sampling-based algorithms that use observation likelihood weighting have shown un…
Many medical decision-making tasks can be framed as partially observed Markov decision processes (POMDPs). However, prevailing two-stage approaches that first learn a POMDP and then solve it often fail because the model that best fits the data may not be well suited for planning. We introduce a new optimization objecti…
Enhances RL in partially observable, noisy environments by uncovering causal states.
problem Making decisions based on incomplete and noisy observations in partially observable Markov decision processes (P2OMDPs). method Causal State Representation under Asynchronous Diffusion Model (CaDiff) framework, incorporating a novel asynchronous diffusion model (ADM) and a new bisimulation metric.
result Enhances returns by at least 14.18% compared to baselines on Roboschool tasks.
This paper formulates a model of utility for a continuous time framework that captures the decision-maker's concern with ambiguity about both the drift and volatility of the driving process. At a technical level, the analysis requires a significant departure from existing continuous time modeling because it cannot be d…
A new method clusters rows of a matrix of point processes.
problem Challenges in analyzing structured point process data.
method Mixture model of multi-level marked point processes, combined with ES algorithm and FPCA.
result An efficient method for clustering rows of a matrix of point processes.
PRL improves off-policy evaluation in partially observed MDPs.
problem Confounding and bias in offline reinforcement learning with unobserved state factors.
method Extends proximal causal inference to POMDPs, identifying and estimating target policy value.
result Semiparametrically efficient estimators for PRL in partially observed MDPs.
Solves ambiguity in incomplete markets by minimizing price measure entropy.
problem Ambiguity in pricing incomplete markets.
method Minimizes the entropy of the price measure from the economic measure, subject to mark-to-market constraints.
result Resolves ambiguity and provides a consistent pricing measure.
The Markov assumption (MA) is fundamental to the empirical validity of reinforcement learning. In this paper, we propose a novel Forward-Backward Learning procedure to test MA in sequential decision making. The proposed test does not assume any parametric form on the joint distribution of the observed data and plays an…
The objective is to study an on-line Hidden Markov model (HMM) estimation-based Q-learning algorithm for partially observable Markov decision process (POMDP) on finite state and action sets. When the full state observation is available, Q-learning finds the optimal action-value function given the current action (Q func…
This work studies the problem of batch off-policy evaluation for Reinforcement Learning in partially observable environments. Off-policy evaluation under partial observability is inherently prone to bias, with risk of arbitrarily large errors. We define the problem of off-policy evaluation for Partially Observable Mark…
Hierarchical clustering has been shown to be valuable in many scenarios. Despite its usefulness to many situations, there is no agreed methodology on how to properly evaluate the hierarchies produced from different techniques, particularly in the case where ground-truth labels are unavailable. This motivates us to prop…
Paper analyzes history-based RL methods for MDPs, introduces a theoretical framework and practical algorithm.
problem Improving RL performance in MDPs using history-based features.
method Theoretical framework for history-based RL, practical algorithm design.
result Practical RL algorithm shows effectiveness on continuous control tasks.
New algorithm proves RL from partial obs is feasible.
problem Difficulty in learning from partial observability.
method Optimism combined with MLE for weakly revealing POMDPs.
result Simple algorithm guarantees polynomial sample efficiency.
New method for evaluating policies in complex decision-making models with hidden variables.
problem Evaluating policies in partially observable Markov decision processes with hidden confounders.
method Introduces novel identification methods and minimax estimation techniques for linking target policy's value and observed data distribution.
result Proposes three estimators for off-policy evaluation in POMDPs with latent confounders, demonstrating their effectiveness through nonasymptotic and asymptotic analysis.
New algorithm solves uncertain Markov decision processes using Wasserstein uncertainty.
problem Solving Markov decision processes with uncertain transition probabilities.
method Distributionally robust Q-learning algorithm for Wasserstein uncertainty. result Convergence of the algorithm proved and demonstrated with real data.
New algorithms learn POMDPs efficiently with hindsight observability.
problem Hardness of learning in POMDPs due to partial observability.
method Hindsight Observable Markov Decision Process (HOMDP) and new algorithms for tabular and function approximation settings.
result Sample-efficient learning in POMDPs with optimal dependence on latent state and observation cardinalities.
New algorithm tackles confounding in offline RL for partially observable MDPs.
problem Confounding in offline reinforcement learning for partially observable MDPs.
method P3O algorithm using proximal causal inference and pessimistic confidence regions.
result Achieves n−1/2-suboptimality for offline RL in POMDPs with confounded data. Optimal policies in Markov decision processes (MDPs) are very sensitive to model misspecification. This raises serious concerns about deploying them in high-stake domains. Robust MDPs (RMDP) provide a promising framework to mitigate vulnerabilities by computing policies with worst-case guarantees in reinforcement learn…
New method optimizes decision-making in uncertain environments.
problem Optimal decision-making under partial observability.
method Nested sequential Monte Carlo algorithm for continuous POMDPs.
result Demonstrated effectiveness on continuous POMDP benchmarks.
This paper is concerned with multi-view reinforcement learning (MVRL), which allows for decision making when agents share common dynamics but adhere to different observation models. We define the MVRL framework by extending partially observable Markov decision processes (POMDPs) to support more than one observation mod…
New framework for reinforcement learning with sporadic state observations.
problem Partial observability in reinforcement learning.
method Action-Triggered Sporadically Traceable Markov Decision Processes (ATST-MDPs).
result Optimistic algorithm achieving regret bound for episodic learning.
Study optimal timing to divest from assets with uncertain future scenarios.
problem Optimal timing to divest from assets with uncertain future scenarios.
method Smooth model of decision making under ambiguity aversion, optimal stopping problem with learning.
result Proves a minimax result reducing the problem to standard optimal stopping problems with learning.
Efficient RL in partially observable risk-sensitive environments with hindsight observations.
problem Risk-sensitive reinforcement learning in partially observable environments.
method Integrates hindsight observations into POMDP framework, develops novel RL algorithm.
result Achieves polynomial regret with provable efficiency, outperforming existing methods.
RL approach for target tracking with unknown dynamics and sensor control.
problem Tracking an unknown target with sensor control.
method Track-MDP formulation for RL, compared with POMDP.
result Optimal RL policy tracks all target paths with certainty.
DRO optimizes decisions under uncertain distributions, considering worst-case scenarios.
problem Optimizing decisions when the distribution of uncertainties is itself uncertain.
method Defines ambiguity sets and seeks decisions optimal under the worst-case distribution.
result DRO models can be connected to regularization techniques and machine learning.