Develops CLTs for Markov chain transition probabilities and policies.
problem Estimating transition probabilities and policies in controlled Markov chains.
method Non-parametric estimator for transition matrices; CLTs for value, Q-, and advantage functions; goodness-of-fit tests.
result Asymptotic normality of estimators under specific logging policies.
Study nonparametric estimator for Markov chain transition matrices in offline setting.
problem Estimating transition matrices of finite controlled Markov chains from logged data.
method Developed sample complexity bounds and conditions for minimaxity.
result Achieving certain statistical risk requires balancing mixing properties and sample size.
Policy gradient algorithm with variable learning rates achieves near-optimal performance in multi-arm bandit problems.
problem Optimizing a policy gradient algorithm for multi-arm bandit problems with variable learning rates.
method Applied Foster-Lyapunov techniques to analyze a Markov chain formed by the state of the algorithm.
result The policy gradient algorithm converges to the optimal arm with logarithmic or poly-logarithmic regret.
The paper introduces a novel method for stable off-policy learning using value function chaining.
problem Stability issues in off-policy reinforcement learning.
method The approach involves learning on-policy first, then chaining off-policy value estimates.
result The method guarantees convergence and can approximate off-policy TD solutions.
Study shows TD(0) with linear approx. converges for reversible Markov chains.
problem TD(0) divergence with off-policy and function approximation.
method Analyzes standard TD(0) with reversible Markov chains, adapting stochastic approximation framework.
result Establishes convergence with probability one for projected Bellman error = 0.
New adaptive temperature selection improves parallel tempering efficiency.
problem Enhancing mixing in multi-modal distributions using parallel tempering.
method Adaptive temperature selection using policy gradient approach.
result Lower integrated autocorrelation times achieved compared to traditional methods.
Bootstrap method for Markov chains in reinforcement learning.
problem Distributional consistency in finite controlled Markov chains with unknown control policies.
method Model-based bootstrap with novel LLN and CLT for visitation counts and transition increments.
result Asymptotically valid confidence intervals for value and Q Q Q -functions in offline RL. This paper investigates methods for estimating the optimal stochastic control policy for a Markov Decision Process with unknown transition dynamics and an unknown reward function. This form of model-free reinforcement learning comprises many real world systems such as playing video games, simulated control tasks, and r…
Algorithm selects optimal experiments in Markov chains to learn unknown quantities.
problem Designing efficient experiments in Markov chains to learn about unknown quantities.
method Proposes extsc{markov-design} algorithm for sequential policy selection.
result Algorithm provably converges to optimal measurement allocation.
Supplier learns to price contracts against a learning retailer.
problem Designing data-driven pricing policies for a supplier facing a learning retailer.
method Connecting to non-stationary online learning, proposing dynamic pricing policies for discrete and continuous demand.
result Supplier's pricing policies lead to sublinear regret bounds under various retailer learning policies.
Study analyzes climate-tech investments across 14 sectors.
problem Accelerating climate-tech innovation in nascent value chains.
method Analysis of 4,172 firms and 12,929 investments over 15 years.
result Only 15% of firms develop end products, 59% support them, and 26% develop services.
We consider the problem of learning a policy for a Markov decision process consistent with data captured on the state-actions pairs followed by the policy. We assume that the policy belongs to a class of parameterized policies which are defined using features associated with the state-action pairs. The features are kno…
This paper develops a stochastic learning-optimization model for resilient automotive supply chains.
problem Supply chain disruptions and volatile demand pose challenges to the UK automotive industry.
method Integrates Bayesian inference with inventory optimization for a two-echelon system subject to stochastic demand and disruptions.
result The integrated approach achieves significant cost reductions and improved resilience during disruptions.
FedSight AI predicts federal funds rate using LLMs and multi-agent reasoning.
problem Predicting Federal Open Market Committee's decisions on federal funds rate.
method Multi-agent framework with large language models, structured and unstructured inputs, and CoD extension for efficient reasoning.
result Achieved 93.75% accuracy and 93.33% stability in predicting FOMC outcomes.
Minimal assumptions analysis of Q-learning with time-varying policies.
problem Finite-time analysis of Q-learning with time-varying policies for discounted MDPs.
method Minimal assumptions, Poisson equation decomposition, sensitivity analysis.
result Established convergence rate and sample complexity for Q-learning.
This paper develops numerical methods for finding optimal dividend pay-out and reinsurance policies. A generalized singular control formulation of surplus and discounted payoff function are introduced, where the surplus is modeled by a regime-switching process subject to both regular and singular controls. To approxima…
SNAPO optimizes policies for complex sequential decisions using differentiable simulation.
problem Optimizing policies for high-dimensional, sequential decisions under uncertainty.
method Embeds neural policy in a differentiable simulator, computes gradients efficiently.
result Produces sensitivities at a cost proportional to one reverse pass, regardless of sensitivity count.
LHIEM model predicts health, income, and employment over years.
problem Lack of path dependency in health policy simulations.
method Discrete-time microsimulation with Markov chain modules.
result Validates health care financing proposal through detailed modeling.
Develops statistical framework for resolving reward function ambiguity in inverse reinforcement learning.
problem Non-uniqueness of reward functions in inverse reinforcement learning.
method Entropy regularization combined with least-squares reconstruction of the reward from the soft Bellman residual.
result Least-squares reward function is unique and consistent with the expert policy.
Study assesses climate risks on supply chains and financial systems using detailed firm emissions data.
problem Lack of firm-level CO2 emissions data hinders assessment of transition risks from carbon pricing.
method Used detailed Hungarian firm emissions data and a simple economic ABM model to simulate carbon pricing impacts.
result 45% of companies are directly exposed to carbon pricing, leading to significant economic and financial losses.
We derive and analyze learning algorithms for apprenticeship learning, policy evaluation, and policy gradient for average reward criteria. Existing algorithms explicitly require an upper bound on the mixing time. In contrast, we build on ideas from Markov chain theory and derive sampling algorithms that do not require …
New method decomposes Markov chain rewards into persistent and transient components.
problem Ambiguity in classical evaluation methods for Markov chains with reducible and periodic states.
method Minimal exact quotient by the real peripheral invariant subspace, decomposing rewards into persistent and transient components.
result Exact comparison with classical methods shows that the new decomposition reallocates the same information, making persistent modes explicit.
In a continuous time stochastic economy, this paper considers the problem of consumption and investment in a financial market in which the representative investor exhibits a change in the discount rate. The investment opportunities are a stock and a riskless account. The market coefficients and discount factor switches…
Study optimal offline RL with uncertainty sets and distribution shifts.
problem Optimal offline reinforcement learning with limited data.
method Construct uncertainty sets and distribution shifts, solve robust Markov decision process.
result Least conservative estimator for unknown true distribution.
We consider the restless Markov bandit problem, in which the state of each arm evolves according to a Markov process independently of the learner's actions. We suggest an algorithm that after T T T steps achieves O ~ ( T ) \tilde{O}(\sqrt{T}) O ~ ( T ) regret with respect to the best policy that knows the distributions of all arms. No ass…
This work is motivated by numerical solutions to Hamilton-Jacobi-Bellman quasi-variational inequalities (HJBQVIs) associated with combined stochastic and impulse control problems. In particular, we consider (i) direct control, (ii) penalized, and (iii) semi-Lagrangian discretization schemes applied to the HJBQVI proble…
Paper provides convergence guarantees for off-policy NAC with finite sample complexity.
problem Convergence analysis of off-policy natural actor-critic algorithm.
method Finite-sample analysis with Importance Sampling and Q-trace algorithm.
result Converges to global optimal policy with sample complexity O ( ε − 3 log 2 ( 1 / ε ) ) \mathcal{O}(ε^{-3}\log^2(1/ε)) O ( ε − 3 log 2 ( 1/ ε )) . Bayesian approach for policy search in stochastic domains.
problem Policy search in stochastic domains.
method Nested probabilistic programs, Lightweight Metropolis-Hastings (LMH) adaptation.
result Similar quality policies learned with simpler algorithm.
Paper analyzes convergence of dynamic policy gradient for MDPs, improving performance in finite-time problems.
problem Optimal policies in finite-time MDPs are not stationary and require epoch-specific training.
method Introduces dynamic policy gradient combining dynamic programming and policy gradient, analyzes convergence for softmax parametrisation.
result Dynamic policy gradient training exploits finite-time structure, leading to better convergence bounds.
New algorithm identifies best policy in MDPs faster.
problem Identifying the best policy in Markov Decision Processes.
method Problem-dependent lower bound and first algorithm with instance-specific sample complexity.
result First algorithm with reduced exploration rate for faster convergence.
This paper tackles data-efficient nonlinear control in Hamiltonian systems using symplectic geometry.
problem Data-efficient nonlinear control in Hamiltonian systems.
method Combines symplectic geometry, recurrence on energy level sets, and chain policies to solve target reachability problems.
result Data requirements depend on geometric and recurrence properties of the Hamiltonian, not the state dimension.
MAGE optimizes policies using action gradients from model-based learning.
problem Lack of direct gradient information from critics in actor-critic methods.
method Model-based actor-critic algorithm that learns action-value gradient.
result MAGE outperforms model-free and model-based baselines on continuous control tasks.
Optimizes consumption under regime-switching economic states with risk-sensitive preferences.
problem Optimizing consumption in an economy with uncertain states and random shocks.
method Risk-sensitive optimization of consumption-utility with a Markov chain model of economic states and i.i.d. random shocks.
result Existence of unique optimal policy and value function in stationary policies.
GenDICE estimates stationary values from offline data.
problem Estimating stationary values from limited offline data.
method Ratio correction based on stationary distribution properties, variational divergence minimization.
result Consistent estimation of stationary values possible in offline settings.
This paper optimizes MDP policies for efficient state aggregation.
problem Optimizing policies in aggregated Markov chains while preserving optimal performance.
method Homomorphic mappings to establish optimal policy equivalence and derive performance bounds.
result Developed HPG and EBHPG methods for efficient aggregation and policy optimization.
Price of anarchy, the performance ratio, which could characterize the loss of efficiency of the distributed supply chain management compared with the integrated supply chain management is discussed by utilizing newsvendor problem in single period which is well-known. In particular, some of remarkable distributed polici…
Novel bounds improve TD learning consistency in RL.
problem Analyzing Temporal Difference learning's performance.
method High-dimensional concentration inequalities and Berry-Esseen bounds for Markov chain induced martingales.
result Sharp high-probability consistency guarantee for TD learning, matching asymptotic variance up to logarithmic factors.
The paper studies how search and distillation improve reasoning in large language models.
problem Improving reasoning capabilities of large language models.
method Viewing chain-of-thought generation as a metastable Markov process, proving benefits of search and distillation.
result Search protocol rewards sparse edges, reducing the expected number of steps to reach different clusters.
Though reinforcement learning has greatly benefited from the incorporation of neural networks, the inability to verify the correctness of such systems limits their use. Current work in explainable deep learning focuses on explaining only a single decision in terms of input features, making it unsuitable for explaining …
Estimates Markov chain variance efficiently without storing samples.
problem Estimating the asymptotic variance of Markov chain functions.
method Linear stochastic approximation of Poisson equation solution.
result Optimal MSE convergence rate with finite sample guarantees.
Paper finds optimal selling rule for pairs trading with stock constraints.
problem Identifying the best time to sell in pairs trading of stocks.
method Optimal pairs-trading selling rule with constraints on trading.
result Closed-form solution for optimal policy determined by a threshold curve.
In this paper we build on previous work which uses inferences techniques, in particular Markov Chain Monte Carlo (MCMC) methods, to solve parameterized control problems. We propose a number of modifications in order to make this approach more practical in general, higher-dimensional spaces. We first introduce a new tar…
This paper studies the potential of the return distribution for exploration in deterministic reinforcement learning (RL) environments. We study network losses and propagation mechanisms for Gaussian, Categorical and Gaussian mixture distributions. Combined with exploration policies that leverage this return distributio…
Backpropagation and the chain rule of derivatives have been prominent; however, the total derivative rule has not enjoyed the same amount of attention. In this work we show how the total derivative rule leads to an intuitive visual framework for creating gradient estimators on graphical models. In particular, previous …
The paper develops RL methods for optimal switching between multiple states.
problem Optimal switching between multiple states in continuous time.
method Entropy-regularized exploration, HJB equations, policy improvement, value function convergence.
result The RL algorithm converges to optimal policies as temperature parameter vanishes.
Market makers optimize bid/ask quotes under hidden Markov chain uncertainty.
problem Optimizing market quotes with hidden factors affecting order intensities.
method Solves stochastic control problem using filtering, control, and PDMPs theory.
result Value function is unique viscosity solution of dynamic programming equation.
Study visualizes actor-critic loss landscapes for inventory optimization.
problem Difficulties in solving multi-store dynamic inventory control problems.
method Low-dimensional visualizations of actor loss function.
result Loss landscapes favor optimal policies in reinforcement learning.
Paper achieves ε − 2 ε^{-2} ε − 2 sample complexity for actor-critic methods with minimal assumptions.
problem Achieving ε − 2 ε^{-2} ε − 2 sample complexity for actor-critic methods under minimal assumptions. method Single-loop, single-timescale implementation; coupled Lyapunov drift framework.
result First i l d e O ( ε − 2 ) ilde{\mathcal{O}}(ε^{-2}) i l d e O ( ε − 2 ) sample complexity guarantee for finding an ε ε ε -optimal policy.