Recent advancements in Distributional Reinforcement Learning (DRL) for modeling loss distributions have shown promise in developing hedging strategies in derivatives markets. A common approach in DRL involves learning the quantiles of loss distributions at specified levels using Quantile Regression (QR). This method is particularly effective in option hedging due to its direct quantile-based risk assessment, such as Value at Risk (VaR) and Conditional Value at Risk (CVaR). However, these risk measures depend on the accurate estimation of extreme quantiles in the loss distribution's tail, which can be imprecise in QR-based DRL due to the rarity and extremity of tail data, as highlighted in the literature. To address this issue, we propose EXtreme DRL (EX-DRL), which enhances extreme quantile prediction by modeling the tail of the loss distribution with a Generalized Pareto Distribution (GPD). This method introduces supplementary data to mitigate the scarcity of extreme quantile observations, thereby improving estimation accuracy through QR. Comprehensive experiments on gamma hedging options demonstrate that EX-DRL improves existing QR-based models by providing more precise estimates of extreme quantiles, thereby improving the computation and reliability of risk metrics for complex financial risk management.
Distributional reinforcement learning improves performance by capturing environmental stochasticity, but a comprehensive theoretical understanding of its effectiveness remains elusive. In addition, the intractable element of the infinite dimensionality of distributions has been overlooked. In this paper, we present a regret analysis of distributional reinforcement learning with general value function approximation in a finite episodic Markov decision process setting. We first introduce a key notion of $\textit{Bellman unbiasedness}$ which is essential for exactly learnable and provably efficient distributional updates in an online manner. Among all types of statistical functionals for representing infinite-dimensional return distributions, our theoretical results demonstrate that only moment functionals can exactly capture the statistical information. Secondly, we propose a provably efficient algorithm, $\texttt{SF-LSVI}$, that achieves a tight regret bound of $\tilde{O}(d_E H^{\frac{3}{2}}\sqrt{K})$ where $H$ is the horizon, $K$ is the number of episodes, and $d_E$ is the eluder dimension of a function class.
We introduce a novel class of algorithms to efficiently approximate the unknown return distributions in policy evaluation problems from distributional reinforcement learning (DRL). The proposed distributional dynamic programming algorithms are suitable for underlying Markov decision processes (MDPs) having an arbitrary probabilistic reward mechanism, including continuous reward distributions with unbounded support being potentially heavy-tailed. For a plain instance of our proposed class of algorithms we prove error bounds, both within Wasserstein and Kolmogorov--Smirnov distances. Furthermore, for return distributions having probability density functions the algorithms yield approximations for these densities; error bounds are given within supremum norm. We introduce the concept of quantile-spline discretizations to come up with algorithms showing promising results in simulation experiments. While the performance of our algorithms can rigorously be analysed they can be seen as universal black box algorithms applicable to a large class of MDPs. We also derive new properties of probability metrics commonly used in DRL on which our quantitative analysis is based.
This paper introduces PG-Rainbow, a novel algorithm that incorporates a distributional reinforcement learning framework with a policy gradient algorithm. Existing policy gradient methods are sample inefficient and rely on the mean of returns when calculating the state-action value function, neglecting the distributional nature of returns in reinforcement learning tasks. To address this issue, we use an Implicit Quantile Network that provides the quantile information of the distribution of rewards to the critic network of the Proximal Policy Optimization algorithm. We show empirical results that through the integration of reward distribution information into the policy network, the policy agent acquires enhanced capabilities to comprehensively evaluate the consequences of potential actions in a given state, facilitating more sophisticated and informed decision-making processes. We evaluate the performance of the proposed algorithm in the Atari-2600 game suite, simulated via the Arcade Learning Environment (ALE).
Risk-sensitive reinforcement learning (RL) is crucial for maintaining reliable performance in high-stakes applications. While traditional RL methods aim to learn a point estimate of the random cumulative cost, distributional RL (DRL) seeks to estimate the entire distribution of it, which leads to a unified framework for handling different risk measures. However, developing policy gradient methods for risk-sensitive DRL is inherently more complex as it involves finding the gradient of a probability measure. This paper introduces a new policy gradient method for risk-sensitive DRL with general coherent risk measures, where we provide an analytical form of the probability measure's gradient for any distribution. For practical use, we design a categorical distributional policy gradient algorithm (CDPG) that approximates any distribution by a categorical family supported on some fixed points. We further provide a finite-support optimality guarantee and a finite-iteration convergence guarantee under inexact policy evaluation and gradient estimation. Through experiments on stochastic Cliffwalk and CartPole environments, we illustrate the benefits of considering a risk-sensitive setting in DRL.
Categorical Distributional Reinforcement Learning (CDRL) has demonstrated superior sample efficiency in learning complex tasks compared to conventional Reinforcement Learning (RL) approaches. However, the practical application of CDRL is encumbered by challenging projection steps, detailed parameter tuning, and domain knowledge. This paper addresses these challenges by introducing a pioneering Continuous Distributional Model-Free RL algorithm tailored for continuous action spaces. The proposed algorithm simplifies the implementation of distributional RL, adopting an actor-critic architecture wherein the critic outputs a continuous probability distribution. Additionally, we propose an ensemble of multiple critics fused through a Kalman fusion mechanism to mitigate overestimation bias. Through a series of experiments, we validate that our proposed method is easy to train and serves as a sample-efficient solution for executing complex continuous-control tasks.
Distributional reinforcement learning (DRL) has achieved empirical success in various domains. One core task in DRL is distributional policy evaluation, which involves estimating the return distribution $\eta^\pi$ for a given policy $\pi$. Distributional temporal difference learning has been accordingly proposed, which extends the classic temporal difference learning (TD) in RL. In this paper, we focus on the non-asymptotic statistical rates of distributional TD. To facilitate theoretical analysis, we propose non-parametric distributional TD (NTD). For a $\gamma$-discounted infinite-horizon tabular Markov decision process, we show that for NTD with a generative model, we need $\tilde{O}(\varepsilon^{-2}\mu_{\min}^{-1}(1-\gamma)^{-3})$ interactions with the environment to achieve an $\varepsilon$-optimal estimator with high probability, when the estimation error is measured by the $1$-Wasserstein. This sample complexity bound is minimax optimal up to logarithmic factors. In addition, we revisit categorical distributional TD (CTD), showing that the same non-asymptotic convergence bounds hold for CTD in the case of the $1$-Wasserstein distance. We also extend our analysis to the more general setting where the data generating process is Markovian. In the Markovian setting, we propose variance-reduced variants of NTD and CTD, and show that both can achieve a $\tilde{O}(\varepsilon^{-2} \mu_{\pi,\min}^{-1}(1-\gamma)^{-3}+t_{mix}\mu_{\pi,\min}^{-1}(1-\gamma)^{-1})$ sample complexity bounds in the case of the $1$-Wasserstein distance, which matches the state-of-the-art statistical results for classic policy evaluation. To achieve the sharp statistical rates, we establish a novel Freedman's inequality in Hilbert spaces. This new Freedman's inequality would be of independent interest for statistical analysis of various infinite-dimensional online learning problems.
In the realm of reinforcement learning (RL), accounting for risk is crucial for making decisions under uncertainty, particularly in applications where safety and reliability are paramount. In this paper, we introduce a general framework on Risk-Sensitive Distributional Reinforcement Learning (RS-DisRL), with static Lipschitz Risk Measures (LRM) and general function approximation. Our framework covers a broad class of risk-sensitive RL, and facilitates analysis of the impact of estimation functions on the effectiveness of RSRL strategies and evaluation of their sample complexity. We design two innovative meta-algorithms: \texttt{RS-DisRL-M}, a model-based strategy for model-based function approximation, and \texttt{RS-DisRL-V}, a model-free approach for general value function approximation. With our novel estimation techniques via Least Squares Regression (LSR) and Maximum Likelihood Estimation (MLE) in distributional RL with augmented Markov Decision Process (MDP), we derive the first $\widetilde{\mathcal{O}}(\sqrt{K})$ dependency of the regret upper bound for RSRL with static LRM, marking a pioneering contribution towards statistically efficient algorithms in this domain.
Transient stability-constrained preventive redispatch plays a crucial role in ensuring power system security and stability. Since redispatch strategies need to simultaneously satisfy complex transient constraints and the economic need, model-based formulation and optimization become extremely challenging. In addition, the increasing uncertainty and variability introduced by renewable sources start to drive the system stability consideration from deterministic to probabilistic, which further exaggerates the complexity. In this paper, a Graph neural network guided Distributional Deep Reinforcement Learning (GD2RL) method is proposed, for the first time, to solve the uncertainty-aware transient stability-constrained preventive redispatch problem. First, a graph neural network-based transient simulator is trained by supervised learning to efficiently generate post-contingency rotor angle curves with the steady-state and contingency as inputs, which serves as a feature extractor for operating states and a surrogate time-domain simulator during the environment interaction for reinforcement learning. Distributional deep reinforcement learning with explicit uncertainty distribution of system operational conditions is then applied to generate the redispatch strategy to balance the user-specified probabilistic stability performance and economy preferences. The full distribution of the post-redispatch transient stability index is directly provided as the output. Case studies on the modified New England 39-bus system validate the proposed method.
This paper contributes a new approach for distributional reinforcement learning which elucidates a clean separation of transition structure and reward in the learning process. Analogous to how the successor representation (SR) describes the expected consequences of behaving according to a given policy, our distributional successor measure (SM) describes the distributional consequences of this behaviour. We formulate the distributional SM as a distribution over distributions and provide theory connecting it with distributional and model-based reinforcement learning. Moreover, we propose an algorithm that learns the distributional SM from data by minimizing a two-level maximum mean discrepancy. Key to our method are a number of algorithmic techniques that are independently valuable for learning generative models of state. As an illustration of the usefulness of the distributional SM, we show that it enables zero-shot risk-sensitive policy evaluation in a way that was not previously possible.
In this paper, we prove that Distributional Reinforcement Learning (DistRL), which learns the return distribution, can obtain second-order bounds in both online and offline RL in general settings with function approximation. Second-order bounds are instance-dependent bounds that scale with the variance of return, which we prove are tighter than the previously known small-loss bounds of distributional RL. To the best of our knowledge, our results are the first second-order bounds for low-rank MDPs and for offline RL. When specializing to contextual bandits (one-step RL problem), we show that a distributional learning based optimism algorithm achieves a second-order worst-case regret bound, and a second-order gap dependent bound, simultaneously. We also empirically demonstrate the benefit of DistRL in contextual bandits on real-world datasets. We highlight that our analysis with DistRL is relatively simple, follows the general framework of optimism in the face of uncertainty and does not require weighted regression. Our results suggest that DistRL is a promising framework for obtaining second-order bounds in general RL settings, thus further reinforcing the benefits of DistRL.
We propose a new algorithm for model-based distributional reinforcement learning (RL), and prove that it is minimax-optimal for approximating return distributions with a generative model (up to logarithmic factors), resolving an open question of Zhang et al. (2023). Our analysis provides new theoretical results on categorical approaches to distributional RL, and also introduces a new distributional Bellman equation, the stochastic categorical CDF Bellman equation, which we expect to be of independent interest. We also provide an experimental study comparing several model-based distributional RL algorithms, with several takeaways for practitioners.
We present a novel statistical approach to incorporating uncertainty awareness in model-free distributional reinforcement learning involving quantile regression-based deep Q networks. The proposed algorithm, $\textit{Calibrated Evidential Quantile Regression in Deep Q Networks (CEQR-DQN)}$, aims to address key challenges associated with separately estimating aleatoric and epistemic uncertainty in stochastic environments. It combines deep evidential learning with quantile calibration based on principles of conformal inference to provide explicit, sample-free computations of $\textit{global}$ uncertainty as opposed to $\textit{local}$ estimates based on simple variance, overcoming limitations of traditional methods in computational and statistical efficiency and handling of out-of-distribution (OOD) observations. Tested on a suite of miniaturized Atari games (i.e., MinAtar), CEQR-DQN is shown to surpass similar existing frameworks in scores and learning speed. Its ability to rigorously evaluate uncertainty improves exploration strategies and can serve as a blueprint for other algorithms requiring uncertainty awareness.
We study distributional off-policy evaluation (OPE), of which the goal is to learn the distribution of the return for a target policy using offline data generated by a different policy. The theoretical foundation of many existing work relies on the supremum-extended statistical distances such as supremum-Wasserstein distance, which are hard to estimate. In contrast, we study the more manageable expectation-extended statistical distances and provide a novel theoretical justification on their validity for learning the return distribution. Based on this attractive property, we propose a new method called Energy Bellman Residual Minimizer (EBRM) for distributional OPE. We provide corresponding in-depth theoretical analyses. We establish a finite-sample error bound for the EBRM estimator under the realizability assumption. Furthermore, we introduce a variant of our method based on a multi-step extension which improves the error bound for non-realizable settings. Notably, unlike prior distributional OPE methods, the theoretical guarantees of our method do not require the completeness assumption.
Growth in the penetration of renewable energy sources makes supply more uncertain and leads to an increase in the system imbalance. This trend, together with the single imbalance pricing, opens an opportunity for balance responsible parties (BRPs) to perform energy arbitrage in the imbalance settlement mechanism. To this end, we propose a battery control framework based on distributional reinforcement learning (DRL). Our proposed control framework takes a risk-sensitive perspective, allowing BRPs to adjust their risk preferences: we aim to optimize a weighted sum of the arbitrage profit and a risk measure while constraining the daily number of cycles for the battery. We assess the performance of our proposed control framework using the Belgian imbalance prices of 2022 and compare two state-of-the-art RL methods, deep Q learning and soft actor-critic. Results reveal that the distributional soft actor-critic method can outperform other methods. Moreover, we note that our fully risk-averse agent appropriately learns to hedge against the risk related to the unknown imbalance price by (dis)charging the battery only when the agent is more certain about the price.
Distributional Reinforcement Learning (RL) estimates return distribution mainly by learning quantile values via minimizing the quantile Huber loss function, entailing a threshold parameter often selected heuristically or via hyperparameter search, which may not generalize well and can be suboptimal. This paper introduces a generalized quantile Huber loss function derived from Wasserstein distance (WD) calculation between Gaussian distributions, capturing noise in predicted (current) and target (Bellman-updated) quantile values. Compared to the classical quantile Huber loss, this innovative loss function enhances robustness against outliers. Notably, the classical Huber loss function can be seen as an approximation of our proposed loss, enabling parameter adjustment by approximating the amount of noise in the data during the learning process. Empirical tests on Atari games, a common application in distributional RL, and a recent hedging strategy using distributional RL, validate the effectiveness of our proposed loss function and its potential for parameter adjustments in distributional RL. The implementation of the proposed loss function is available here.
We propose a novel algorithmic framework for distributional reinforcement learning, based on learning finite-dimensional mean embeddings of return distributions. We derive several new algorithms for dynamic programming and temporal-difference learning based on this framework, provide asymptotic convergence theory, and examine the empirical performance of the algorithms on a suite of tabular tasks. Further, we show that this approach can be straightforwardly combined with deep reinforcement learning, and obtain a new deep RL agent that improves over baseline distributional approaches on the Arcade Learning Environment.
Generally, Reinforcement Learning (RL) agent updates its policy by repetitively interacting with the environment, contingent on the received rewards to observed states and undertaken actions. However, the environmental disturbance, commonly leading to noisy observations (e.g., rewards and states), could significantly shape the performance of agent. Furthermore, the learning performance of Multi-Agent Reinforcement Learning (MARL) is more susceptible to noise due to the interference among intelligent agents. Therefore, it becomes imperative to revolutionize the design of MARL, so as to capably ameliorate the annoying impact of noisy rewards. In this paper, we propose a novel decomposition-based multi-agent distributional RL method by approximating the globally shared noisy reward by a Gaussian mixture model (GMM) and decomposing it into the combination of individual distributional local rewards, with which each agent can be updated locally through distributional RL. Moreover, a diffusion model (DM) is leveraged for reward generation in order to mitigate the issue of costly interaction expenditure for learning distributions. Furthermore, the optimality of the distribution decomposition is theoretically validated, while the design of loss function is carefully calibrated to avoid the decomposition ambiguity. We also verify the effectiveness of the proposed method through extensive simulation experiments with noisy rewards. Besides, different risk-sensitive policies are evaluated in order to demonstrate the superiority of distributional RL in different MARL tasks.
While distributional reinforcement learning (DistRL) has been empirically effective, the question of when and why it is better than vanilla, non-distributional RL has remained unanswered. This paper explains the benefits of DistRL through the lens of small-loss bounds, which are instance-dependent bounds that scale with optimal achievable cost. Particularly, our bounds converge much faster than those from non-distributional approaches if the optimal cost is small. As warmup, we propose a distributional contextual bandit (DistCB) algorithm, which we show enjoys small-loss regret bounds and empirically outperforms the state-of-the-art on three real-world tasks. In online RL, we propose a DistRL algorithm that constructs confidence sets using maximum likelihood estimation. We prove that our algorithm enjoys novel small-loss PAC bounds in low-rank MDPs. As part of our analysis, we introduce the $\ell_1$ distributional eluder dimension which may be of independent interest. Then, in offline RL, we show that pessimistic DistRL enjoys small-loss PAC bounds that are novel to the offline setting and are more robust to bad single-policy coverage.
Reinforcement learning (RL) has a rich history in neuroscience, from early work on dopamine as a reward prediction error signal (Schultz et al., 1997) to recent work proposing that the brain could implement a form of 'distributional reinforcement learning' popularized in machine learning (Dabney et al., 2020). There has been a close link between theoretical advances in reinforcement learning and neuroscience experiments throughout this literature, and the theories describing the experimental data have therefore become increasingly complex. Here, we provide an introduction and mathematical background to many of the methods that have been used in systems neroscience. We start with an overview of the RL problem and classical temporal difference algorithms, followed by a discussion of 'model-free', 'model-based', and intermediate RL algorithms. We then introduce deep reinforcement learning and discuss how this framework has led to new insights in neuroscience. This includes a particular focus on meta-reinforcement learning (Wang et al., 2018) and distributional RL (Dabney et al., 2020). Finally, we discuss potential shortcomings of the RL formalism for neuroscience and highlight open questions in the field. Code that implements the methods discussed and generates the figures is also provided.