Study shows accurate OPE depends on calibrated behaviour policy models.
problem Estimating a behaviour policy for OPE when true policy is unknown.
method Empirical studies comparing parametric vs non-parametric models.
result Simple non-parametric k-nearest neighbors model produces better calibrated behaviour policy estimates.
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.
A new policy switching technique improves offline RL performance.
problem Challenges in adapting off-policy algorithms to different datasets and tasks.
method Combines off-policy RL and BC, using epistemic uncertainty for policy switching.
result Outperforms individual algorithms and state-of-the-art methods on benchmarks.
Assessment of risk levels for existing credit accounts is important to the implementation of bank policies and offering financial products. This paper uses cluster analysis of behaviour of credit card accounts to help assess credit risk level. Account behaviour is modelled parametrically and we then implement the behav…
SPIBB improves safe policy improvement with a softer baseline approach.
problem Improving policies safely in reinforcement learning.
method Baseline Bootstrapping algorithm (SPIBB) that allows policy search over a wider set of policies, controlling policy change according to local model uncertainty.
result Significant improvement in safe policy improvement over existing methods.
Improves BC policies by generating new plausible trajectories.
problem Sub-optimal data quality in BC leads to poor policy performance.
method Trajectory Stitching (TS) generates new plausible transitions.
result TS significantly improves behavioural policies over original data.
The paper proposes a new policy training objective to reduce exploration in self-play.
problem Training policies to mimic MCTS search behavior can lead to excessive exploration.
method Derive a policy gradient expression using MCTS value estimates to reduce exploration.
result Empirically evaluated policies show reduced exploration compared to MCTS-based training.
CODA resolves coordination issues in offline multi-agent reinforcement learning.
problem Coordination failure in offline multi-agent reinforcement learning.
method Diffusion-based multi-agent trajectory generator for data augmentation.
result CODA resolves coordination pathologies in continuous polynomial games and complex benchmarks.
Entangled bisimulation improves policy learning from visual input.
problem Learning generalizeable policies from visual input in the presence of visual distractions.
method Proposes entangled bisimulation, a bisimulation metric for continuous state and action spaces.
result Entangled bisimulation improves policy learning on the Distracting Control Suite (DCS).
Improves RL from historical data by stitching trajectories.
problem Lack of high-quality data for offline RL.
method Trajectory Stitching (TS) to augment historical data with synthetic actions.
result Improves RL policy performance over baseline.
In this work, we take a fresh look at some old and new algorithms for off-policy, return-based reinforcement learning. Expressing these in a common form, we derive a novel algorithm, Retrace(λ), with three desired properties: (1) it has low variance; (2) it safely uses samples collected from any behaviour policy, wha…
Paper uses Apprenticeship Learning to model player behavior in interactive narratives.
problem Understanding and simulating player behavior in interactive narratives.
method Receding Horizon IRL (RHIRL) to learn reward functions and policies.
result RHIRL can learn action sequences and generate behavior similar to specific players.
Develops a method to define admissible rewards for robust policy evaluation in RL.
problem Defining a reward function for robust off-policy evaluation in RL with limited data.
method Identifies an admissible set of reward functions ensuring policies are close to past behavior and can be evaluated with high confidence.
result Demonstrates the approach on synthetic and real-world domains, including a critical care application.
The paper improves dialogue quality estimation using a novel user satisfaction model.
problem Improving dialogue quality estimation in spoken dialogue systems.
method Proposes a novel user satisfaction estimator based on BiLSTMs and reinforcement learning.
result The novel user satisfaction estimator outperforms previous models in terms of user satisfaction and task success.
Improved multi-step TD learning with control variates reduces variance and improves performance.
problem Variance in multi-step TD learning causes divergence in off-policy settings.
method Per-decision control variates for multi-step TD algorithms.
result Control variates significantly improve performance in both on and off-policy tasks.
New methods improve robust decision-making under uncertainty in off-policy evaluation.
problem Statistical uncertainty and causal considerations in off-policy evaluation.
method Marginal Ratio (MR) estimator, Conformal Off-Policy Prediction (COPP), causal bounds.
result Improved robustness and uncertainty quantification in off-policy decision-making.
Develops a generic two-layer framework for adaptive ABMs.
problem Bi-level adaptation problem in ABMs: agents adapt to environment, and environment adapts to agents.
method Formalizes bi-level problem as a Stackelberg game with conditional policies, solving coupled non-linear equations.
result Unified framework for adaptive ABMs, addressing traditional ABM limitations.
MO2 learns useful behaviours from past experience for new tasks.
problem Discovering useful behaviours from past experience and transferring them to new tasks.
method Model-Based Offline Options (MO2) framework supporting sample-efficient bottleneck option discovery over continuous state-action spaces.
result MO2 outperforms recent option learning methods on complex long-horizon continuous control tasks.
Studies report that firms do not invest in cost-effective green technologies. While economic barriers can explain parts of the gap, behavioural aspects cause further under-valuation. This could be partly due to systematic deviations of decision-making agents' perceptions from normative benchmarks, and partly due to the…
Improved RL policies from offline data with relaxed BC constraints.
problem Overestimation bias in offline RL due to lack of interaction with environment.
method Introducing a policy constraint via behavioural cloning (BC) and adjusting the balance between RL and BC.
result Refined policies outperform baseline and match/exceed complex alternatives.
Strategic behaviour is one of the main explanations for cost overruns. It can theoretically be supported by agency theory, in which strategic behaviour is the result of asymmetric information between the principal and agent. This paper gives a formal account of this relation by a signalling game. This is a game with in…
Diffusion models mimic human actions in sequential tasks.
problem Cloning human behavior in dynamic environments is challenging.
method Adapting diffusion models to handle stochastic, multimodal, and correlated actions.
result Diffusion models closely replicate human behavior in robotic and gaming tasks.
This paper explores how AI systems can learn moral behavior from economic entities.
problem Achieving moral behavior in AI systems.
method Analyses the analogy between machine learning and economic entities.
result Implicit specifications may work better than explicit ones for AI problems.
In many real-world settings, a team of agents must coordinate their behaviour while acting in a decentralised way. At the same time, it is often possible to train the agents in a centralised fashion in a simulated or laboratory setting, where global state information is available and communication constraints are lifte…
MAP-Elites generates diverse trading strategies for improved execution performance.
problem Optimizing trading execution schedules in volatile market conditions.
method Quality-diversity algorithm (MAP-Elites) generating a portfolio of specialized strategies.
result Diverse strategies achieve 8-10% performance improvements, validating quality-diversity methods.
QMIX combines per-agent values to create decentralised policies.
problem Training decentralised policies from centralised learning.
method QMIX uses a mixing network to estimate joint action-values as a monotonic combination of per-agent values.
result QMIX significantly outperforms existing methods on the StarCraft Multi-Agent Challenge (SMAC).
A new theorem and algorithm solve off-policy policy gradient problems.
problem Solving the theoretical gap in off-policy policy gradient methods.
method Introduced an off-policy policy gradient theorem using emphatic weightings and developed the ACE algorithm.
result Demonstrated ACE finds the optimal solution in off-policy learning, unlike previous methods.
Protects proprietary policies from imitation learning by training adversarial policy ensembles.
problem Protecting policies from external observers cloning them.
method Introduces a reinforcement learning framework that trains an ensemble of near-optimal policies, making demonstrations useless for external observers.
result Demonstrates the existence of 'non-clonable' ensembles and provides a solution to the optimization problem.
Automated scoring prioritizes risky driving behavior in telematic auto insurance policies.
problem Identifying risky driving behavior in telematic auto insurance policies using machine learning.
method Bayesian approach using MCMC to model propensity of policyholders to undertake trips resulting in positive classification.
result The approach improves efficiency of human resource allocation in identifying risky driving behavior.
This work refines imitation learning to approximate optimal state-feedback policies for a quadcopter model.
problem Designing optimal control policies for complex systems like quadcopters.
method Supervised imitation learning using deep neural networks trained on optimal trajectories.
result Deep neural networks can approximate optimal state-feedback policies with high accuracy, even with two layers.
We introduce a methodology for efficiently computing a lower bound to empowerment, allowing it to be used as an unsupervised cost function for policy learning in real-time control. Empowerment, being the channel capacity between actions and states, maximises the influence of an agent on its near future. It has been sho…
Paper analyzes convergence of entropy-regularized reinforcement learning methods.
problem Ensuring sub-optimality control in entropy-regularized RL.
method Unified analysis of regularized MPI and VI schemes, providing convergence rates.
result Established sufficient conditions for convergence and provided explicit rates.
AI agent plays CSGO deathmatch with human-like style.
problem Lack of API for CSGO limits data for reinforcement learning.
method Behavioural cloning on large noisy and expert datasets.
result Matches human difficulty level in deathmatch mode.
Developed an explainable DRL model for financial portfolio management.
problem Inability of DRL agents to provide interpretable financial investment policies.
method Integrating PPO with feature importance techniques (SHAP, LIME) to enhance transparency.
result Ability to interpret DRL agent actions in prediction time.
Demonstration-regularized RL reduces sample complexity for policy identification.
problem Improving reinforcement learning's sample efficiency with expert demonstrations.
method KL-regularization of behavior cloning using expert demonstrations.
result Demonstration-regularized RL achieves optimal policy identification with reduced sample complexity.
Theory for RLHF generalization under reward shift and clipped KL.
problem Theoretical understanding of RLHF generalization, especially with reward shift and clipped KL.
method Developed generalization theory for RLHF, accounting for reward shift and clipped KL.
result Presented generalization bounds for RLHF, suggesting generalization error from sampling, reward shift, and KL clipping.
Paper improves off-policy evaluation by estimating behavior policy.
problem Evaluating policies with data from a different behavior policy.
method Importance sampling with an estimated behavior policy.
result Estimating behavior policy reduces mean squared error.
Estimates Schwarzian derivative on long complex projective tubes.
problem Behaviour of Schwarzian derivative on complex projective structures.
method Analyzes Schwarzian derivative on long complex projective tubes, estimating its pairing with infinitesimal earthquakes and graftings.
result Obtains bounds for the variation of renormalized volume under complex earthquake paths and its asymptotic behavior under pinching.
Bayesian framework inserts prior knowledge in RL for faster task solving.
problem Faster transfer learning across reinforcement learning tasks.
method Bayesian posterior distribution combining task-specific and prior knowledge.
result Significant speed ups achieved in maze solving.
Method uses ANN to estimate incentive salience from large behavioral data.
problem Estimating incentive salience in naturalistic settings.
method Artificial Neural Networks (ANNs) for latent state approximation.
result ANNs produce better representations for predicting future behaviour.
Study tackles balancing policy switching costs in offline RL.
problem Balancing the cost of policy switching in offline RL.
method Optimal transport ideas and Net Actor-Critic algorithm.
result Demonstrated efficiency on multiple RL benchmarks.
New methods estimate policy value and gradients for deterministic policies from off-policy data.
problem Estimating policy value and gradients for deterministic policies from off-policy data.
method Proposed new doubly robust estimators based on kernelization approaches.
result Demonstrated a rate independent of horizon length for policy value and gradient estimation.
New method estimates state-action stationary distribution for better off-policy policy evaluation.
problem Accurately estimating state-action stationary distribution for off-policy policy evaluation.
method Estimated Mixture Policy (EMP) for state and state-action stationary distribution corrections.
result Empirical validation shows improved accuracy over state-of-the-art methods.
The paper explains why estimating a history-dependent policy can reduce MSE in reinforcement learning.
problem Understanding why history-dependent policies can improve MSE in off-policy evaluation.
method The paper derives a bias-variance decomposition of MSE for various OPE estimators, showing how history-dependent policies can decrease variance and increase bias.
result History-dependent policies can decrease the variance of importance sampling estimators, leading to lower MSE.
Regularizes model-based planning using energy-based models for efficient learning.
problem Challenges in using learned dynamics models for accurate planning.
method Regularization using energy estimates of state transitions.
result Proposes effective regularization method for planning with pre-trained dynamics models.
New method improves off-policy critic evaluation in reinforcement learning.
problem High variance and instability in off-policy policy evaluation.
method Doubly robust estimators applied to actor-critic algorithms.
result Doubly robust estimation significantly improves performance in continuous control tasks.
Safe RL for autonomous vehicles using PCPO with trust regions and parallel learners.
problem Unexplainable behaviours and lack of safety guarantees in RL for real vehicles.
method PCPO framework with trust regions and parallel learners.
result Safe learning confirmed for autonomous vehicles with fast convergence.
A new estimator for evaluating policies in unknown environments.
problem Evaluating policies when both logging policy and value function are unknown.
method Doubly-Robust (DR) off-policy evaluation (OPE) estimator, DRUnknown, that estimates both the logging policy and value function.
result DRUnknown achieves the smallest asymptotic variance and is optimal when both models are correctly specified.