Kernel-UCBVI algorithm balances exploration and exploitation in metric state-action spaces.
problem Exploration-exploitation dilemma in finite-horizon reinforcement learning with metric state-action spaces.
method Kernel-UCBVI, leveraging smoothness and kernel estimators of rewards and transitions.
result First regret bound for kernel-based RL using smoothing kernels, O(H3K2d/(2d+1)). ZoomRL learns efficient strategies for large state-action spaces using a metric.
problem Handling large state-action spaces in reinforcement learning.
method ZoomRL leverages continuous bandits to adaptively discretize the joint space.
result Achieves worst-case regret of $ ilde{O}(H^{ rac{5}{2}} K^{ rac{d+1}{d+2}})$.
KeRNS tackles non-stationary reinforcement learning in metric spaces.
problem Non-stationary reinforcement learning in metric spaces.
method KeRNS uses time-dependent kernels to model non-stationary Markov Decision Processes (MDPs).
result KeRNS achieves a regret bound that scales with the covering dimension and total variation of the MDP.
Model-free Reinforcement Learning (RL) algorithms such as Q-learning [Watkins, Dayan 92] have been widely used in practice and can achieve human level performance in applications such as video games [Mnih et al. 15]. Recently, equipped with the idea of optimism in the face of uncertainty, Q-learning algorithms [Jin, Al…
We present an efficient algorithm for model-free episodic reinforcement learning on large (potentially continuous) state-action spaces. Our algorithm is based on a novel Q-learning policy with adaptive data-driven discretization. The central idea is to maintain a finer partition of the state-action space in regions w…
A new method estimates multi-dimensional value distributions using Hilbert space embeddings.
problem Estimating value distributions in complex, multi-dimensional reinforcement learning settings.
method Hilbert space mappings and kernel mean embeddings to estimate the kernel mean embedding of multi-dimensional value distributions.
result Uniform convergence guarantees and robust off-policy evaluation demonstrated in simulations.
New RL method handles large state-action spaces with complex models.
problem Complex models and large state-action spaces in reinforcement learning.
method π-KRVI, an optimistic modification of least-squares value iteration using kernel ridge regression.
result First order-optimal regret guarantees under general settings, improving over state of the art.
New RL method reduces sample complexity for large state-action spaces.
problem Handling large state-action spaces in RL with general Q-functions.
method Nonparametric Q-learning using kernel ridge regression.
result Sample complexity is order optimal with respect to ε and kernel complexity.
Novel framework proves fast RL convergence in continuous spaces.
problem Analyzing stability in continuous state-action RL.
method Introduces a novel framework to analyze stability properties of RL.
result Highlights two key stability properties and demonstrates their satisfaction in RL.
A new RL paradigm reduces state-action-value function approximation inefficiency.
problem Challenges in state-action-value function approximation for RL.
method State Action Separable Reinforcement Learning (sasRL) decouples action space from value function learning.
result sasRL achieves up to 75% better performance than state-of-the-art MDP-based RL algorithms.
This paper introduces CENIE to quantify environment novelty for better UED.
problem Challenges in measuring environment novelty for effective UED.
method CENIE framework using state-action space coverage and Gaussian Mixture Models.
result CENIE improves UED performance across multiple benchmarks.
We study online reinforcement learning for finite-horizon deterministic control systems with {\it arbitrary} state and action spaces. Suppose that the transition dynamics and reward function is unknown, but the state and action space is endowed with a metric that characterizes the proximity between different states and…
New method for natural policy gradients converges linearly.
problem Improving natural policy gradient methods for better convergence.
method Fisher-Rao gradient flow applied to state-action distributions.
result Linear convergence rate with geometry-dependent factor.
We develop a normative framework for hierarchical model-based policy optimization based on applying second-order methods in the space of all possible state-action paths. The resulting natural path gradient performs policy updates in a manner which is sensitive to the long-range correlational structure of the induced st…
We tackle the Multi-task Batch Reinforcement Learning problem. Given multiple datasets collected from different tasks, we train a multi-task policy to perform well in unseen tasks sampled from the same distribution. The task identities of the unseen tasks are not provided. To perform well, the policy must infer the tas…
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.
New RL approach uses future state and action visitation measures for better exploration.
problem Improving exploration in reinforcement learning.
method Intrinsic reward based on future state and action visitation measures, using contraction operators.
result Policies achieve good state-action space coverage and high performance.
New method recovers diverse policies from expert data using state-action pair weighting.
problem Recovering diverse policies from expert trajectories.
method Pointwise mutual information weighted behavioral cloning.
result Effective in focusing on state-action pairs most representative of the style.
Paper introduces SALE for better state-action learning in RL.
problem Challenges in representation learning for low-level states in RL.
method Introduces SALE, a novel approach for learning embeddings of state-action interactions.
result TD7 algorithm significantly outperforms existing continuous control algorithms.
The paper proposes a principle for dynamically adjusting the granularity of reinforcement learning abstractions.
problem Lack of general principles for dynamically adjusting the granularity of reinforcement learning abstractions.
method The paper proposes a principle based on rate-distortion theory, formalized through a performance certificate decomposing value error into learning and abstraction error bounds.
result Soft state-action abstractions can achieve near-optimal performance under substantial lossy compression of state and action information.
This work tackles model-based RL by optimizing state-action queries to learn policies with minimal data.
problem Expensive state transitions in practical RL problems limit the use of standard RL algorithms.
method Bayesian optimal experimental design to guide selection of state-action queries.
result Data-efficient RL approach that learns optimal policies with up to 1,000x less data.
A new framework for offline RL improves policy flexibility and regularity.
problem Lack of environmental interactions in offline RL leads to poor policy performance.
method Proposes a behavior-regularized implicit policy framework with modified policy-matching methods.
result The framework improves policy effectiveness and robustness beyond static datasets.
Estimation of importance sampling weights for off-policy evaluation of contextual bandits often results in imbalance - a mismatch between the desired and the actual distribution of state-action pairs after weighting. In this work we present balanced off-policy evaluation (B-OPE), a generic method for estimating weights…
Federated Q-learning achieves linear speedup with heterogeneity, improving sample complexity.
problem Collaborative learning in distributed RL settings with limited data sharing.
method Analyzes synchronous and asynchronous federated Q-learning, proposes importance averaging.
result Achieves linear speedup with heterogeneity, robust to local trajectory heterogeneity.
Paper tackles covariate shift in offline IL using less proficient behavior data.
problem Mitigating covariate shift in Imitation Learning with limited data coverage.
method Model-based IL from Offline Data (MILO) framework.
result MILO can combat covariate shift even with sub-optimal behavior policy data.
New framework for RL transfer learning with state-action mismatch.
problem High sample complexity in RL from scratch.
method Embeddings to transfer knowledge between MDPs with different state- and action-spaces.
result Successful transfer learning in scenarios with state- and action-space mismatches.
Develops a dynamic mean field theory for reinforcement learning.
problem Finite state and action Bayesian reinforcement learning in large state spaces.
method Analogies with statistical physics, interpreting probabilities as couplings and values as spins, solving mean field equations.
result State-action values are statistically independent in the asymptotic state space limit, with exact or approximate equations for computation.
GUM tackles MARL by avoiding overestimation through state-marginal restriction.
problem Overestimation of values in large joint state-action spaces.
method Greedy UnMixing through state-marginal restriction and unmixing.
result Superior performance compared to existing Q-learning and general MARL algorithms.
Develops a new reinforcement learning framework for complex control problems.
problem Continuous-time extended mean field control with deterministic policies.
method Model-free sensitivity formula, deterministic policy gradient, local value and advantage-rate representations.
result Demonstrates efficiency, stability, and robustness in solving complex control problems.
Most real-world problems have huge state and/or action spaces. Therefore, a naive application of existing tabular solution methods is not tractable on such problems. Nonetheless, these solution methods are quite useful if an agent has access to a relatively small state-action space homomorphism of the true environment …
A new method, Count-MORL, improves offline reinforcement learning by using state-action frequency.
problem Improving offline reinforcement learning performance.
method Integrates count-based conservatism into model-based offline reinforcement learning.
result The learned policy is near-optimal and outperforms existing methods.
A critical and challenging problem in reinforcement learning is how to learn the state-action value function from the experience replay buffer and simultaneously keep sample efficiency and faster convergence to a high quality solution. In prior works, transitions are uniformly sampled at random from the replay buffer o…
New findings reveal discount regularization can be seen as a strong prior, leading to poor performance in unevenly sampled data.
problem Discount regularization leads to poor performance in unevenly sampled data.
method Equivalence theorem showing discount regularization as a strong prior, setting regularization parameters locally for individual state-action pairs.
result Discount regularization can be seen as a strong prior, leading to poor performance in unevenly sampled data.
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.
RPO uses past and future state-action info for better policy optimization.
problem Sample inefficiency in on-policy reinforcement learning methods.
method Reflective Policy Optimization (RPO) integrates past and future state-action info for policy improvement.
result RPO improves policy performance and contracts the solution space, leading to faster convergence.
The paper improves Q-learning by incorporating pessimism for better sample efficiency.
problem Improving sample efficiency in asynchronous Q-learning with non-i.i.d. data.
method Developed an algorithmic framework that incorporates the principle of pessimism into asynchronous Q-learning, penalizing infrequently-visited state-action pairs based on suitable lower confidence bounds (LCBs).
result Achieved near-optimal sample complexity, providing theoretical support for the use of pessimism in non-i.i.d. data.
Adaptive discretization improves model-based RL in large spaces.
problem Efficient model-based reinforcement learning in large state-action spaces.
method Optimistic one-step value iteration with adaptive discretization.
result Adaptive discretization leads to better performance and lower memory usage.
Leveraging an equivalence property in the state-space of a Markov Decision Process (MDP) has been investigated in several studies. This paper studies equivalence structure in the reinforcement learning (RL) setup, where transition distributions are no longer assumed to be known. We present a notion of similarity betwee…
Model-free learning for multi-agent stochastic games is an active area of research. Existing reinforcement learning algorithms, however, are often restricted to zero-sum games, and are applicable only in small state-action spaces or other simplified settings. Here, we develop a new data efficient Deep-Q-learning method…
Hierarchical Modular Reinforcement Learning (HMRL), consists of 2 layered learning where Profit Sharing works to plan a prey position in the higher layer and Q-learning method trains the state-actions to the target in the lower layer. In this paper, we expanded HMRL to multi-target problem to take the distance between …
RandQL is a new model-free algorithm for MDPs with a novel learning rate randomization approach.
problem Minimizing regret in episodic MDPs with model-free methods.
method Randomized Q-learning with learning rate randomization.
result RandQL achieves optimal regret bounds in both tabular and metric state-action spaces.
Real-world applications require RL algorithms to act safely. During learning process, it is likely that the agent executes sub-optimal actions that may lead to unsafe/poor states of the system. Exploration is particularly brittle in high-dimensional state/action space due to increased number of low-performing actions. …
Consider a Markov decision process (MDP) that admits a set of state-action features, which can linearly express the process's probabilistic transition model. We propose a parametric Q-learning algorithm that finds an approximate-optimal policy using a sample size proportional to the feature dimension K and invariant …
The paper introduces a new intrinsic reward method for exploration in reinforcement learning.
problem Improving exploration in reinforcement learning agents.
method Intrinsic rewards proportional to the entropy of future state-action features.
result The new objective leads to improved visitation of features within individual trajectories.
This paper unifies three regularization methods in batch reinforcement learning.
problem Learning overly-complex models in batch reinforcement learning.
method Unified weighted average transition matrix framework for three regularization methods.
result Empirical evaluation confirms intuitions about regularization methods' performance.
BRPO optimizes batch RL policies to better exploit state-action differences.
problem Batch RL's conservatism limits exploitation of state-action differences.
method Proposes residual policies and derives BRPO to maximize policy performance.
result BRPO achieves state-of-the-art performance in various tasks.
Model-based reinforcement learning algorithms make decisions by building and utilizing a model of the environment. However, none of the existing algorithms attempts to infer the dynamics of any state-action pair from known state-action pairs before meeting it for sufficient times. We propose a new model-based method ca…
SPAQL improves RL by adaptively partitioning state-action space and learning a time-invariant policy.
problem Efficient model-free reinforcement learning with scalable algorithms.
method Adaptive Q-learning with UCB and Boltzmann exploration, automatically tuning temperature.
result SPAQL converges faster and uses fewer resources than AQL, showing higher sample efficiency.