Active Screening for Recurrent Diseases: A Reinforcement Learning Approach

The flying sidekick traveling salesman problem with stochastic travel time: A reinforcement learning approach

P2P power trading based on reinforcement learning for nanogrid clusters

Introduction Reinforcement learning formalizes the concept of learning from interactions.1 Broadly, reinforcement learning focuses on a setting in which an agent (decision maker) sequentially interacts with an environment that is partially unknown to them. At each stage, the agent takes an action and receives a reward. The objective of the agent is to maximize rewards accumulated in the long run. There are many situations in health care where decisions are made sequentially for which reinforcement learning approaches could prove useful for decision making. Throughout this article, we consider treatment prescription as an archetypical example to connect reinforcement learning concepts to a health care setting. In this setting, the care provider, the prescribed treatment, and the patients can be viewed as the agent, the action, and the environment, respectively, as depicted in Figure 1.Figure 1: Sequential treatment of AKI or CKD complications modeled as a reinforcement learning problem.Background In this section, with the objective of making reinforcement learning literature more accessible to a clinical audience, we briefly introduce related fundamental concepts and approaches. We refer the interested reader to Sutton and Barto1 for a comprehensive introduction to reinforcement learning. Markov Decision Processes Markov decision processes (MDPs) are a formalism of the sequential decision-making problem that has been central to the theoretical and practical advancements of reinforcement learning. In each stage of an MDP, the agent observes the state of the environment and takes an action, which, in turn, results in a change of the state. This change of state is assumed to be probabilistic with the next state being determined only by the preceding state, the chosen action, and the transition probability. The agent also receives a reward that is a function of the taken action, the preceding state, and the subsequent state. In an MDP, the objective of the agent is to maximize the return defined as the reward accumulated over a time horizon. In some applications, it is common to consider the horizon to be infinite, in which case the future rewards are discounted by a factor smaller than one. The selection of action by the agent on the basis of the observed state is known as the policy. More formally, a policy is a probabilistic mapping from states to each possible action. Because the policy and the reward are a function of the state, it is critical to estimate the utility of being in a certain state. More specifically, the value function is defined as the expected return starting from a given state under the chosen policy. Under this formalism, the objective of the agent is to find the optimal policy that maximizes the value function for all states. Reinforcement Learning Methods Action-value methods are a class of reinforcement learning methods in which the actions are chosen on the basis of the estimation of their long-term value. A prominent example of an action-value method is Q-learning in which the agent iteratively takes actions with the highest estimated values and updates the action-state value function on the basis of new observations. Policy gradient methods are another class of reinforcement learning methods that seek to optimize the policy directly instead of choosing actions on the basis of their respective estimated value. Such methods could be advantageous in health care applications that entail a large number of possible actions, e.g., when recommending a wide range of drug dosages or treatment options. Clinical Applications Reinforcement learning frameworks and methods are broadly applicable to clinical settings in which decisions are made sequentially. A prominent clinical application of reinforcement learning is for treatment recommendation, which has been studied across a variety of diseases and treatments including radiation and chemotherapy for cancer, brain stimulation for epilepsy, and treatment strategies for sepsis.2–5 In such treatment recommendation settings, a policy is commonly known as a dynamic treatment regime. There are various other clinical applications of reinforcement learning including diagnosis, medical imaging, and decision support tools (see refs. 2–5 and the references therein). Reinforcement Learning in Nephrology Although there have been recent applications of machine learning in nephrology,6,7 to the best of the authors' knowledge, the application of reinforcement learning to nephrology has been primarily limited to optimizing the erythropoietin dosage in hemodialysis patients.8,9 However, there are other settings where reinforcement learning has the potential to improve patient care in nephrology. For example, reinforcement learning methods can be adopted in the treatment of the complications of AKI or CKD (Figure 1). In this problem, the state models the conditions of the patient (e.g., vital signs, laboratory test results including urine and blood tests, and urine output measurements). The action refers to the treatment options (e.g., the dosage of medications such as sodium polystyrene sulfonate, and hemodialysis). The reward models the improvement in patient conditions. Similarly, reinforcement learning can help automate and optimize the dosage of immunosuppressive drugs in kidney transplants. Challenges and Opportunities Despite the success of reinforcement learning in several simplified clinical settings, their large-scale application to patient care faces several open challenges. The complexity of human biology complicates modeling clinical decision making as a reinforcement learning problem. The state space in such settings is often enormous, which could make a purely computational approach infeasible. Moreover, modeling all potential objectives a priori as a reward function may not be feasible. To overcome these challenges and realize the potential of reinforcement learning, clinical insight can play a pivotal role. More specifically, restricting the state space to only include highly relevant clinical variables could greatly reduce the computational complexity. Furthermore, using inverse reinforcement learning,2 relevant reward functions can be learned from retrospective studies assuming the optimality of clinical decisions. Another critical challenge is addressing moral and ethical concerns. It is imperative to ensure that reinforcement learning methods do not cause harm to the patient. To this end, there exists a need for a thorough validation of such methods before their use in patient care. Hence, there is a need to go beyond retrospective studies that have been used for the proof of concept of most existing reinforcement learning methods in health care applications.2,3 The lessons learned from the success of reinforcement learning in other application areas (e.g., self-driving cars) can help navigate the path to realizing its potential in health care. Accessible open-source simulation environments that enable researchers to compare various approaches are essential to the field of reinforcement learning. OpenAI Gym is currently the leading toolkit containing a wide range of simulated environments, e.g., surgical robotics.10 The development of high-quality and reliable simulation environments for nephrology and other health care applications can facilitate the development and validation of reinforcement learning methods beyond limited retrospective studies. The adoption of methods validated in such simulation environments in actual clinical settings will require clinicians' oversight. Similar to how self-driving cars require a human driver to ensure collision avoidance, clinicians' oversight is critical to ensure the safety of the patients, especially in the early stages of the adoption of reinforcement learning methods. The data from clinicians' decisions (e.g., overruling the automated treatment recommendation) can be used to improve the reliability of autonomous systems over time and reduce the burden of clinicians' oversight.

Cloud storage is a target of advanced persistent threats (APTs), where a sophisticated adversary attempts to steal sensitive data in a continuous manner. Human monitoring and intervention is an integral part of the reinforcement learning (RL) approaches to defend against APTs. In this paper, prospect theory (PT) is used to model the subjective behavior of the cloud storage defender in assigning computing resources (processing units) to scan and monitor the cloud storage system against an APT attacker bot, which attempts to steal information from the cloud. Under a constraint on the total number of processing units and a lack of knowledge of the opponent's resource allocation strategy, we study the defense performance of a federated maximum-likelihood deep Q-network (FMLDQ) RL algorithm against a sophisticated branching dueling deep Q-network (BDQ) RL attack algorithm. Specifically, the RL strategy for the defender is affected by subjective decisions in estimating the processing units of the attacker. Simulation results show that when the defender has more resources than the attacker, an EUT-based defense strategy (without human intervention) yields better data protection. On the other hand, when the defender has fewer resources, a PT-based defense strategy (with human intervention) is better.

To find the path that minimizes the time to navigate between two given points in a fluid flow is known as Zermelo's problem. Here, we investigate it by using a Reinforcement Learning (RL) approach for the case of a vessel that has a slip velocity with fixed intensity, Vs, but variable direction and navigating in a 2D turbulent sea. We show that an Actor-Critic RL algorithm is able to find quasioptimal solutions for both time-independent and chaotically evolving flow configurations. For the frozen case, we also compared the results with strategies obtained analytically from continuous Optimal Navigation (ON) protocols. We show that for our application, ON solutions are unstable for the typical duration of the navigation process and are, therefore, not useful in practice. On the other hand, RL solutions are much more robust with respect to small changes in the initial conditions and to external noise, even when Vs is much smaller than the maximum flow velocity. Furthermore, we show how the RL approach is able to take advantage of the flow properties in order to reach the target, especially when the steering speed is small.

Deep reinforcement learning (RL) is achieving significant success in various applications like control, robotics, games, resource management, and scheduling. However, the important problem of emergency evacuation, which clearly could benefit from RL, has been largely unaddressed. Indeed, emergency evacuation is a complex task that is difficult to solve with RL. An emergency situation is highly dynamic, with a lot of changing variables and complex constraints that make it challenging to solve. Also, there is no standard benchmark environment available that can be used to train RL agents for evacuation. A realistic environment can be complex to design. In this article, we propose the first fire evacuation environment to train RL agents for evacuation planning. The environment is modeled as a graph capturing the building structure. It consists of realistic features like fire spread, uncertainty, and bottlenecks. The implementation of our environment is in the OpenAI gym format, to facilitate future research. We also propose a new RL approach that entails pretraining the network weights of a DQN-based agent [DQN/Double-DQN (DDQN)/Dueling-DQN] to incorporate information on the shortest path to the exit. We achieved this by using tabular <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula> -learning to learn the shortest path on the building model’s graph. This information is transferred to the network by deliberately overfitting it on the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula> -matrix. Then, the pretrained DQN model is trained on the fire evacuation environment to generate the optimal evacuation path under time varying conditions due to fire spread, bottlenecks, and uncertainty. We perform comparisons of the proposed approach with state-of-the-art RL algorithms like DQN, DDQN, Dueling-DQN, PPO, VPG, state-action-reward-state-action (SARSA), actor–critic method, and ACKTR. The results show that our method is able to outperform state-of-the-art models by a huge margin including the original DQN-based models. Finally, our model is tested on a large and complex real building consisting of 91 rooms, with the possibility to move to any other room, hence giving 8281 actions. In order to reduce the action space, we propose a strategy that involves one step simulation. That is, an action importance vector is added to the final output of the pretrained DQN and acts like an attention mechanism. Using this strategy, the action space is reduced by 90.1%. In this manner, the model is able to deal with large action spaces. Hence, our model achieves near optimal performance on the real world emergency environment.

Estimation bias is an important index for evaluating the performance of reinforcement learning (RL) algorithms. The popular RL algorithms, such as Q -learning and deep Q -network (DQN), often suffer overestimation due to the maximum operation in estimating the maximum expected action values of the next states, while double Q -learning (DQ) and double DQN may fall into underestimation by using a double estimator (DE) to avoid overestimation. To keep the balance between overestimation and underestimation, we propose a novel integrated DE (IDE) architecture by combining the maximum operation and DE operation to estimate the maximum expected action value. Based on IDE, two RL algorithms: 1) integrated DQ (IDQ) and 2) its deep network version, that is, integrated double DQN (IDDQN), are proposed. The main idea of the proposed RL algorithms is that the maximum and DE operations are integrated to eliminate the estimation bias, where one estimator is stochastically used to perform action selection based on the maximum operation, and the convex combination of two estimators is used to carry out action evaluation. We theoretically analyze the reason of estimation bias caused by using nonmaximum operation to estimate the maximum expected value and investigate the possible reasons of underestimation existence in DQ. We also prove the unbiasedness of IDE and convergence of IDQ. Experiments on the grid world and Atari 2600 games indicate that IDQ and IDDQN can reduce or even eliminate estimation bias effectively, enable the learning to be more stable and balanced, and improve the performance effectively.

Multimedia content in 5G/6G networks makes safe, confidential, and efficient content delivery difficult. Intelligent systems that adapt to the ever‐changing network environment are needed to distribute multimedia content in these networks. Reinforcement learning (RL) can optimize multimedia content distribution based on network congestion, capacity, and user preferences. This study proposes RL‐based intelligent multimedia content distribution. RL algorithms learn from the network environment and generate optimum judgments incorporating several aspects of the suggested framework. The framework delivers multimedia material securely and privately with great quality. This study provides an intelligent multimedia content delivery architecture that uses RL approaches to solve 5G/6G content delivery problems. This research presents an RL system optimized with the double DQN algorithm having a reward of 51604.93 in 7000 episodes for efficient video file sharing on intracity buses. The RL agent balances network congestion and bandwidth by leveraging multiple sources such as bus and intersection caches and base stations, improving secure multimedia content delivery in 5G/6G networks and enhancing the passenger experience. The study confirms the system's effectiveness using reward and loss metrics and identifies potential future research directions. Future work could explore additional RL algorithms, scalability for larger networks, complex delivery scenarios, and integration with blockchain and edge computing for improved security and efficiency in multimedia content delivery.

Reinforcement learning (RL) approaches have recently been spotlighted for use in adaptive traffic-signal control on an area-wide level. Most researchers have employed multi-agent reinforcement learning (MARL) algorithms wherein each agent shares a holistic traffic state and cooperates with other agents to reach a common goal. However, MARL algorithms cannot guarantee a global optimal solution unless the actions of all agents are fully coordinated. The present study employs a RL algorithm that recognizes an entire traffic state and jointly controls all the traffic signals of multiple intersections. With this approach, a deep Q-network (DQN) that depends solely on traffic images is extended to overcome the curse of dimensionality that is associated with a large state and action space. Several front layers in a deep convolutional neural network (CNN) to approximate the true Q-function are shared by each intersection approach. Weight parameters connecting the last hidden layer to the output layer are fixed. The proposed methodology outperforms a fixed-signal operation, a fully actuated signal operation, a multi-agent RL control without coordination, and a multi-agent RL control with partial coordination.

Collaborative optimization of multi-energy multi-microgrid system: A hierarchical trust-region multi-agent reinforcement learning approach

Influence maximization (IM) aims to find a set of seed nodes in a social network that maximizes the influence spread. While most IM problems focus on classical influence cascades (e.g., Independent Cascade and Linear Threshold) which assume individual influence cascade probability is independent of the number of neighbors, recent studies by sociologists show that many influence cascades follow a pattern called complex contagion (CC), where influence cascade probability is much higher when more neighbors are influenced. Nonetheless, there are very limited studies on complex contagion influence maximization (CCIM) problems. This is partly because CC is non-submodular, the solution of which has been an open challenge. In this study, we propose the first reinforcement learning (RL) approach to CCIM. We find that a key obstacle in applying existing RL approaches to CCIM is the reward sparseness issue, which comes from two distinct sources. We then design a new RL algorithm that uses the CCIM problem structure to address the issue. Empirical results show that our approach achieves the state-of-the-art performance on four real-world networks.

This paper proposes a reinforcement learning (RL) algorithm for the security problem of state estimation of cyber-physical system (CPS) under denial-of-service (DoS) attacks. The security of CPS will inevitably decline when faced with malicious cyber attacks. In order to analyze the impact of cyber attacks on CPS performance, a Kalman filter, as an adaptive state estimation technology, is combined with an RL method to evaluate the issue of system security, where estimation performance is adopted as an evaluation criterion. Then, the transition of estimation error covariance under a DoS attack is described as a Markov decision process, and the RL algorithm could be applied to resolve the optimal countermeasures. Meanwhile, the interactive combat between defender and attacker could be regarded as a two-player zero-sum game, where the Nash equilibrium policy exists but needs to be solved. Considering the energy constraints, the action selection of both sides will be restricted by setting certain cost functions. The proposed RL approach is designed from three different perspectives, including the defender, the attacker and the interactive game of two opposite sides. In addition, the framework of Q-learning and state–action–reward–state–action (SARSA) methods are investigated separately in this paper to analyze the influence of different RL algorithms. The results show that both algorithms obtain the corresponding optimal policy and the Nash equilibrium policy of the zero-sum interactive game. Through comparative analysis of two algorithms, it is verified that the differences between Q-Learning and SARSA could be applied effectively into the secure state estimation in CPS.

Inventory management remains a cornerstone of effective supply chain performance, directly influencing cost efficiency, service quality, and organizational agility. In today’s hypercompetitive and uncertain market environment, inventory decisions must account for complex variables such as fluctuating demand, supply disruptions, lead time variability, and market seasonality. Traditional inventory control models such as the Economic Order Quantity (EOQ), base-stock policies, and (s, S) strategies are often static in nature. They rely on pre-defined parameters and assume stationarity in demand and supply, limiting their ability to respond dynamically to real-time changes. In contrast, Reinforcement Learning (RL) offers a paradigm shift in how inventory decisions can be optimized. As a subfield of machine learning, RL enables agents to learn optimal strategies through repeated interactions with an environment, using trial-and-error exploration and reward-based feedback. RL agents can observe the system state (e.g., inventory levels, demand signals, lead time status), choose actions (e.g., place an order or wait), and receive feedback in the form of rewards (e.g., service level achievements or cost penalties), thus iteratively improving their policies. This study explores how RL can be applied to optimize inventory management in environments characterized by uncertainty and real-time decision-making needs. Specifically, we investigate how different RL algorithms such as Q-learning, Deep Q Networks (DQN), and Policy Gradient methods perform in various inventory scenarios. Additionally, we examine the computational and operational implications of deploying RL in real-world settings, including issues of model convergence, exploration-exploitation tradeoffs, data requirements, and scalability. We also discuss how RL can complement other AI techniques such as demand forecasting models and predictive analytics in creating end-to-end intelligent supply chain solutions. By bridging the gap between theoretical RL frameworks and practical inventory management applications, this paper contributes to both the academic literature and industrial practice. Our goal is to demonstrate that RL is not only a theoretically elegant solution but also a viable tool for achieving inventory efficiency and supply chain resilience.

Dynamic pricing has become a critical capability in modern digital commerce platforms, enabling organizations to optimize revenue, profit margins, and inventory turnover in rapidly changing market environments. Traditional rule-based and statistical pricing models often fail to adapt effectively to complex and dynamic consumer behavior. This paper explores the application of reinforcement learning (RL) techniques for dynamic pricing within Microsoft Dynamics 365 Commerce. The study formulates the pricing process as a Markov Decision Process (MDP), where pricing decisions are treated as sequential actions that influence future demand and revenue outcomes. Various reinforcement learning approaches, including contextual multi-armed bandits and Deep Q-Networks, are evaluated for their suitability in enterprise retail environments. The proposed framework integrates RL models with the existing pricing engine of Microsoft Dynamics 365 Commerce using API-based overrides and cloud-based machine learning services. Experimental evaluation demonstrates that reinforcement learning-driven pricing strategies outperform static and rule-based pricing methods in terms of revenue uplift, margin optimization, and inventory efficiency. The paper also discusses implementation challenges such as regulatory compliance, fairness considerations, cold-start problems, and real-time system constraints. The findings suggest that reinforcement learning can provide a scalable and adaptive pricing solution for enterprise commerce systems, offering significant competitive advantage in digital retail ecosystems.

Reinforcement learning (RL) algorithms have been successfully applied to control tasks associated with unmanned aerial vehicles and robotics. In recent years, safe RL has been proposed to allow the safe execution of RL algorithms in industrial and mission-critical systems that operate in closed loops. However, if the system operating conditions change, such as when an unknown fault occurs, typical safe RL algorithms cannot adapt while retaining past knowledge. Continual RL algorithms have been proposed to address this issue. However, the impact of continual adaptation on the system’s safety is an understudied problem. In this paper, we study the intersection of safe and continual RL. First, we empirically demonstrate that a popular continual RL algorithm, elastic weight consolidation, does not satisfy safety constraints in nonlinear systems subject to varying operating conditions. Specifically, we study the MuJoCo HalfCheetah and Ant environments with velocity constraints and sudden joint loss non-stationarity. Then, we show that an agent trained using constrained policy optimization, a safe RL algorithm, experiences catastrophic forgetting in continual learning settings. With this in mind, we explore a simple reward-shaping method to ensure that elastic weight consolidation prioritizes remembering both safety and task performance for safety-constrained, nonlinear, and non-stationary systems.