The worldwide economy and environmental sustainability depend on eff icient and reliable supply chains, in which container shipping plays a crucial role as an environmentally friendly mode of transport. Liner shipping companies seek to improve operational efficiency by solving the stowage planning problem. Due to many complex combinatorial aspects, stowage planning is challenging and often decomposed into two NP-hard subproblems: master and slot planning. This article proposes AI2STOW, an end-to-end deep reinforcement learning model with feasibility projection and an action mask to create master plans under demand uncertainty with global objectives and constraints, including paired block stowage patterms. Our experimental results demonstrate that AI2STOW outperforms baseline methods from reinforcement learning and stochastic programming in objective performance and computational efficiency, based on simulated instances reflecting the scale of realistic vessels and operational planning horizons.
In this paper, we investigate a joint source-channel encoding (JSCE) scheme in an intelligent reflecting surface (IRS)-assisted multi-user semantic communication system. Semantic encoding not only compresses redundant information, but also enhances information orthogonality in a semantic feature space. Meanwhile, the IRS can adjust the spatial orthogonality, enabling concurrent multi-user semantic communication in densely deployed wireless networks to improve spectrum efficiency. We aim to maximize the users' semantic throughput by jointly optimizing the users' scheduling, the IRS's passive beamforming, and the semantic encoding strategies. To tackle this non-convex problem, we propose an explainable deep neural network-driven deep reinforcement learning (XD-DRL) framework. Specifically, we employ a deep neural network (DNN) to serve as a joint source-channel semantic encoder, enabling transmitters to extract semantic features from raw images. By leveraging structural similarity, we assign some DNN weight coefficients as the IRS's phase shifts, allowing simultaneous optimization of IRS's passive beamforming and DNN training. Given the IRS's passive beamforming and semantic encoding strategies, user scheduling is optimized using the DRL method. Numerical results validate that our JSCE scheme achieves superior semantic throughput compared to the conventional schemes and efficiently reduces the semantic encoder's mode size in multi-user scenarios.
Physics-informed machine learning (PIML) provides a promising solution for building energy modeling and can serve as a virtual environment to enable reinforcement learning (RL) agents to interact and learn. However, challenges remain in efficiently integrating physics priors, evaluating the effectiveness of physics constraints, balancing model accuracy and physics consistency, and enabling real-world implementation. To address these gaps, this study introduces a Physics-Informed Modularized Neural Network (PI-ModNN), which incorporates physics priors through a physics-informed model structure, loss functions, and hard constraints. A new evaluation metric called "temperature response violation" is developed to quantify the physical consistency of data-driven building dynamic models under varying control inputs and training data sizes. Additionally, a physics prior evaluation framework based on rule importance is proposed to assess the contribution of each individual physics prior, offering guidance on selecting appropriate PIML techniques. Results indicate that incorporating physical priors does not always improve model performance; inappropriate priors may decrease model accuracy and consistency. However, hard constraints are effective in enforcing model consistency. Furthermore, we present a general workflow for developing control-oriented PIML models and integrating them with deep reinforcement learning (DRL). Following this framework, a case study implementing DRL in an office space over three months demonstrates potential energy savings of 31.4%. Finally, we provide a general guideline for integrating data-driven models with advanced building control through a four-step evaluation framework, paving the way for reliable and scalable deployment of advanced building controls.
Automated control of personalized multiple anesthetics in clinical Total Intravenous Anesthesia (TIVA) is crucial yet challenging. Current systems, including target-controlled infusion (TCI) and closed-loop systems, either rely on relatively static pharmacokinetic/pharmacodynamic (PK/PD) models or focus on single anesthetic control, limiting personalization and collaborative control. To address these issues, we propose a novel framework, Value Decomposition Multi-Agent Deep Reinforcement Learning (VD-MADRL). VD-MADRL optimizes the collaboration between two anesthetics propofol (Agent I) and remifentanil (Agent II). And It uses a Markov Game (MG) to identify optimal actions among heterogeneous agents. We employ various value function decomposition methods to resolve the credit allocation problem and enhance collaborative control. We also introduce a multivariate environment model based on random forest (RF) for anesthesia state simulation. Additionally, a data resampling and alignment technique ensures synchronized trajectory data. Our experiments on general and thoracic surgery datasets show that VD-MADRL performs better than human experience. It improves dose precision and keeps anesthesia states stable, providing great clinical value.
The application of deep reinforcement learning algorithms to economic battery dispatch problems has significantly increased recently. However, optimizing battery dispatch over long horizons can be challenging due to delayed rewards. In our experiments we observe poor performance of popular actor-critic algorithms when trained on yearly episodes with hourly resolution. To address this, we propose an approach extending soft actor-critic (SAC) with learning from demonstrations. The special feature of our approach is that, due to the absence of expert demonstrations, the demonstration data is generated through simple, rule-based policies. We conduct a case study on a grid-connected microgrid and use if-then-else statements based on the wholesale price of electricity to collect demonstrations. These are stored in a separate replay buffer and sampled with linearly decaying probability along with the agent's own experiences. Despite these minimal modifications and the imperfections in the demonstration data, the results show a drastic performance improvement regarding both sample efficiency and final rewards. We further show that the proposed method reliably outperforms the demonstrator and is robust to the choice of rule, as long as the rule is sufficient to guide early training into the right direction.
Online 3D Bin Packing Problem (3D-BPP) has widespread applications in industrial automation. Existing methods usually solve the problem with limited resolution of spatial discretization, and/or cannot deal with complex practical constraints well. We propose to enhance the practical applicability of online 3D-BPP via learning on a novel hierarchical representation, packing configuration tree (PCT). PCT is a full-fledged description of the state and action space of bin packing which can support packing policy learning based on deep reinforcement learning (DRL). The size of the packing action space is proportional to the number of leaf nodes, making the DRL model easy to train and well-performing even with continuous solution space. We further discover the potential of PCT as tree-based planners in deliberately solving packing problems of industrial significance, including large-scale packing and different variations of BPP setting. A recursive packing method is proposed to decompose large-scale packing into smaller sub-trees while a spatial ensemble mechanism integrates local solutions into global. For different BPP variations with additional decision variables, such as lookahead, buffering, and offline packing, we propose a unified planning framework enabling out-of-the-box problem solving. Extensive evaluations demonstrate that our method outperforms existing online BPP baselines and is versatile in incorporating various practical constraints. The planning process excels across large-scale problems and diverse problem variations. We develop a real-world packing robot for industrial warehousing, with careful designs accounting for constrained placement and transportation stability. Our packing robot operates reliably and efficiently on unprotected pallets at 10 seconds per box. It achieves averagely 19 boxes per pallet with 57.4% space utilization for relatively large-size boxes.
This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible. Multi-agent task allocation (MATA) plays a vital role in cooperative multi-agent systems, with significant implications for applications such as logistics, search and rescue, and robotic coordination. Although traditional deep reinforcement learning (DRL) methods have been shown to be promising, their effectiveness is hindered by a reliance on manually designed reward functions and inefficiencies in dynamic environments. In this paper, an inverse reinforcement learning (IRL)-based framework is proposed, in which multi-head self-attention (MHSA) and graph attention mechanisms are incorporated to enhance reward function learning and task execution efficiency. Expert demonstrations are utilized to infer optimal reward densities, allowing dependence on handcrafted designs to be reduced and adaptability to be improved. Extensive experiments validate the superiority of the proposed method over widely used multi-agent reinforcement learning (MARL) algorithms in terms of both cumulative rewards and task execution efficiency.
Dynamic hedging is a financial strategy that consists in periodically transacting one or multiple financial assets to offset the risk associated with a correlated liability. Deep Reinforcement Learning (DRL) algorithms have been used to find optimal solutions to dynamic hedging problems by framing them as sequential decision-making problems. However, most previous work assesses the performance of only one or two DRL algorithms, making an objective comparison across algorithms difficult. In this paper, we compare the performance of eight DRL algorithms in the context of dynamic hedging; Monte Carlo Policy Gradient (MCPG), Proximal Policy Optimization (PPO), along with four variants of Deep Q-Learning (DQL) and two variants of Deep Deterministic Policy Gradient (DDPG). Two of these variants represent a novel application to the task of dynamic hedging. In our experiments, we use the Black-Scholes delta hedge as a baseline and simulate the dataset using a GJR-GARCH(1,1) model. Results show that MCPG, followed by PPO, obtain the best performance in terms of the root semi-quadratic penalty. Moreover, MCPG is the only algorithm to outperform the Black-Scholes delta hedge baseline with the allotted computational budget, possibly due to the sparsity of rewards in our environment.
This paper presents Post-Decision Proximal Policy Optimization (PDPPO), a novel variation of the leading deep reinforcement learning method, Proximal Policy Optimization (PPO). The PDPPO state transition process is divided into two steps: a deterministic step resulting in the post-decision state and a stochastic step leading to the next state. Our approach incorporates post-decision states and dual critics to reduce the problem's dimensionality and enhance the accuracy of value function estimation. Lot-sizing is a mixed integer programming problem for which we exemplify such dynamics. The objective of lot-sizing is to optimize production, delivery fulfillment, and inventory levels in uncertain demand and cost parameters. This paper evaluates the performance of PDPPO across various environments and configurations. Notably, PDPPO with a dual critic architecture achieves nearly double the maximum reward of vanilla PPO in specific scenarios, requiring fewer episode iterations and demonstrating faster and more consistent learning across different initializations. On average, PDPPO outperforms PPO in environments with a stochastic component in the state transition. These results support the benefits of using a post-decision state. Integrating this post-decision state in the value function approximation leads to more informed and efficient learning in high-dimensional and stochastic environments.
Dealing with environments with sparse rewards has always been crucial for systems developed to operate in autonomous open-ended learning settings. Intrinsic Motivations could be an effective way to help Deep Reinforcement Learning algorithms learn in such scenarios. In fact, intrinsic reward signals, such as novelty or curiosity, are generally adopted to improve exploration when extrinsic rewards are delayed or absent. Building on previous works, we tackle the problem of learning policies in the presence of sparse rewards by proposing a two-level architecture that alternates an ''intrinsically driven'' phase of exploration and autonomous sub-goal generation, to a phase of sparse reward, goal-directed policy learning. The idea is to build several small networks, each one specialized on a particular sub-path, and use them as starting points for future exploration without the need to further explore from scratch previously learnt paths. Two versions of the system have been trained and tested in the Gym SuperMarioBros environment without considering any additional extrinsic reward. The results show the validity of our approach and the importance of autonomously segment the environment to generate an efficient path towards the final goal.
A novel approach to detecting anomalies in time series data is presented in this paper. This approach is pivotal in domains such as data centers, sensor networks, and finance. Traditional methods often struggle with manual parameter tuning and cannot adapt to new anomaly types. Our method overcomes these limitations by integrating Deep Reinforcement Learning (DRL) with a Variational Autoencoder (VAE) and Active Learning. By incorporating a Long Short-Term Memory (LSTM) network, our approach models sequential data and its dependencies effectively, allowing for the detection of new anomaly classes with minimal labeled data. Our innovative DRL- VAE and Active Learning combination significantly improves existing methods, as shown by our evaluations on real-world datasets, enhancing anomaly detection techniques and advancing time series analysis.
Grasping large flat objects, such as books or keyboards lying horizontally, presents significant challenges for single-arm robotic systems, often requiring extra actions like pushing objects against walls or moving them to the edge of a surface to facilitate grasping. In contrast, dual-arm manipulation, inspired by human dexterity, offers a more refined solution by directly coordinating both arms to lift and grasp the object without the need for complex repositioning. In this paper, we propose a model-free deep reinforcement learning (DRL) framework to enable dual-arm coordination for grasping large flat objects. We utilize a large-scale grasp pose detection model as a backbone to extract high-dimensional features from input images, which are then used as the state representation in a reinforcement learning (RL) model. A CNN-based Proximal Policy Optimization (PPO) algorithm with shared Actor-Critic layers is employed to learn coordinated dual-arm grasp actions. The system is trained and tested in Isaac Gym and deployed to real robots. Experimental results demonstrate that our policy can effectively grasp large flat objects without requiring additional maneuvers. Furthermore, the policy exhibits strong generalization capabilities, successfully handling unseen objects. Importantly, it can be directly transferred to real robots without fine-tuning, consistently outperforming baseline methods.
Deep reinforcement learning agents, trained on raw pixel inputs, often fail to generalize beyond their training environments, relying on spurious correlations and irrelevant background details. To address this issue, object-centric agents have recently emerged. However, they require different representations tailored to the task specifications. Contrary to deep agents, no single object-centric architecture can be applied to any environment. Inspired by principles of cognitive science and Occam's Razor, we introduce Object-Centric Attention via Masking (OCCAM), which selectively preserves task-relevant entities while filtering out irrelevant visual information. Specifically, OCCAM takes advantage of the object-centric inductive bias. Empirical evaluations on Atari benchmarks demonstrate that OCCAM significantly improves robustness to novel perturbations and reduces sample complexity while showing similar or improved performance compared to conventional pixel-based RL. These results suggest that structured abstraction can enhance generalization without requiring explicit symbolic representations or domain-specific object extraction pipelines.
Log messages record important system runtime information and are useful for detecting anomalous behaviors and managing modern software systems. Many supervised and unsupervised learning methods have been proposed recently for log-based anomaly detection. State-of-the-art unsupervised methods predict the next log event given a log sequence and apply fixed configurations that use the same filter condition (i.e. k, the top k predicted log events will be regarded as normal next events) which leads to inferior performance in the detection stage because it sets one fixed k for all log sequences, which ignores the dynamic nature and variance in different log sequences. Recently, deep reinforcement learning (DRL) are widely applied to make intelligent decisions in a dynamic environment. In this work, we contend that it is necessary to apply adaptive filters for different log sequences. To achieve this, we propose a novel approach based on DRL to construct a learned adaptive filter and apply different normal/abnormal filter thresholds for different log sequences. We define the Markov Decision Process (MDP) and formulate the learned adaptive filter as a problem that can be solved by DRL. We evaluate the learned adaptive filter on two state-of-the-art log-based anomaly detection unsupervised approaches DeepLog and LogAnomaly in two datasets HDFS and BGL. Extensive experiments show that our approach outperforms the fixed configurations and achieves significantly better performance in log-based anomaly detection.
Path planning and optimization for unmanned aerial vehicles (UAVs)-assisted next-generation wireless networks is critical for mobility management and ensuring UAV safety and ubiquitous connectivity, especially in dense urban environments with street canyons and tall buildings. Traditional statistical and model-based techniques have been successfully used for path optimization in communication networks. However, when dynamic channel propagation characteristics such as line-of-sight (LOS), interference, handover, and signal-to-interference and noise ratio (SINR) are included in path optimization, statistical and model-based path planning solutions become obsolete since they cannot adapt to the dynamic and time-varying wireless channels, especially in the mmWave bands. In this paper, we propose a novel model-free actor-critic deep reinforcement learning (AC-DRL) framework for path optimization in UAV-assisted 5G mmWave wireless networks, which combines four important aspects of UAV communication: \textit{flight time, handover, connectivity and SINR}. We train an AC-RL agent that enables a UAV connected to a gNB to determine the optimal path to a desired destination in the shortest possible time with minimal gNB handover, while maintaining connectivity and the highest possible SINR. We train our model with data from a powerful ray tracing tool called Wireless InSite, which uses 3D images of the propagation environment and provides data that closely resembles the real propagation environment. The simulation results show that our system has superior performance in tracking high SINR compared to other selected RL algorithms.
To address the challenge of insufficient interactivity and behavioral diversity in autonomous driving decision-making, this paper proposes a Cognitive Hierarchical Agent for Reasoning and Motion Stylization (CHARMS). By leveraging Level-k game theory, CHARMS captures human-like reasoning patterns through a two-stage training pipeline comprising reinforcement learning pretraining and supervised fine-tuning. This enables the resulting models to exhibit diverse and human-like behaviors, enhancing their decision-making capacity and interaction fidelity in complex traffic environments. Building upon this capability, we further develop a scenario generation framework that utilizes the Poisson cognitive hierarchy theory to control the distribution of vehicles with different driving styles through Poisson and binomial sampling. Experimental results demonstrate that CHARMS is capable of both making intelligent driving decisions as an ego vehicle and generating diverse, realistic driving scenarios as environment vehicles. The code for CHARMS is released at https://github.com/chuduanfeng/CHARMS.
This thesis delves into the forefront of wireless communication by exploring the synergistic integration of three transformative technologies: STAR-RIS, CoMP, and NOMA. Driven by the ever-increasing demand for higher data rates, improved spectral efficiency, and expanded coverage in the evolving landscape of 6G development, this research investigates the potential of these technologies to revolutionize future wireless networks. The thesis analyzes the performance gains achievable through strategic deployment of STAR-RIS, focusing on mitigating inter-cell interference, enhancing signal strength, and extending coverage to cell-edge users. Resource sharing strategies for STAR-RIS elements are explored, optimizing both transmission and reflection functionalities. Analytical frameworks are developed to quantify the benefits of STAR-RIS assisted CoMP-NOMA networks under realistic channel conditions, deriving key performance metrics such as ergodic rates and outage probabilities. Additionally, the research delves into energy-efficient design approaches for CoMP-NOMA networks incorporating RIS, proposing novel RIS configurations and optimization algorithms to achieve a balance between performance and energy consumption. Furthermore, the application of Deep Reinforcement Learning (DRL) techniques for intelligent and adaptive optimization in aerial RIS-assisted CoMP-NOMA networks is explored, aiming to maximize network sum rate while meeting user quality of service requirements. Through a comprehensive investigation of these technologies and their synergistic potential, this thesis contributes valuable insights into the future of wireless communication, paving the way for the development of more efficient, reliable, and sustainable networks capable of meeting the demands of our increasingly connected world.
Integrated into existing Mobile Edge Computing (MEC) systems, Unmanned Aerial Vehicles (UAVs) serve as a cornerstone in meeting the stringent requirements of future Internet of Things (IoT) networks. The current endeavor studies an MEC system, in which a computationally-empowered UAV, wirelessly linked to a cloud server, is destined for task offloading in uplink transmission of IoT devices. The performance of this system is studied by formulating a resource allocation problem, which aims to maximize the long-term computed task efficiency, while ensuring the stability of task buffers at the IoT devices, UAV and cloud. The problem jointly optimizes the uplink transmit power of IoT devices and their offloading decisions, the trajectory of the UAV and computing power at all transceivers. Regarding the non-convex and stochastic nature of the problem, we devise a multi-step solution approach. Initially, by invoking the fractional programming and Lyapunov theory, we transform the long-term optimization problem into an equivalent per-time-slot form. Subsequently, we recast the reformulated problem as a Markov Decision Process (MDP), which reflects the network dynamics. The MDP model, eventually, serves for training a Meta Twin Delayed Deep Deterministic Policy Gradient (MTD3) agent, in charge of adaptive resource allocation with respect to the MEC system variations derived from the mobility of the UAV and IoT devices. Simulations reveal the dominance of our proposed resource allocation approach over its Deep Reinforcement Learning (DRL)-powered counterparts, increasing computed task efficiency and reducing task buffer lengths.
The economic feasibility of nuclear microreactors will depend on minimizing operating costs through advancements in autonomous control, especially when these microreactors are operating alongside other types of energy systems (e.g., renewable energy). This study explores the application of deep reinforcement learning (RL) for real-time drum control in microreactors, exploring performance in regard to load-following scenarios. By leveraging a point kinetics model with thermal and xenon feedback, we first establish a baseline using a single-output RL agent, then compare it against a traditional proportional-integral-derivative (PID) controller. This study demonstrates that RL controllers, including both single- and multi-agent RL (MARL) frameworks, can achieve similar or even superior load-following performance as traditional PID control across a range of load-following scenarios. In short transients, the RL agent was able to reduce the tracking error rate in comparison to PID. Over extended 300-minute load-following scenarios in which xenon feedback becomes a dominant factor, PID maintained better accuracy, but RL still remained within a 1% error margin despite being trained only on short-duration scenarios. This highlights RL's strong ability to generalize and extrapolate to longer, more complex transients, affording substantial reductions in training costs and reduced overfitting. Furthermore, when control was extended to multiple drums, MARL enabled independent drum control as well as maintained reactor symmetry constraints without sacrificing performance -- an objective that standard single-agent RL could not learn. We also found that, as increasing levels of Gaussian noise were added to the power measurements, the RL controllers were able to maintain lower error rates than PID, and to do so with less control effort.
As rapidly growing AI computational demands accelerate the need for new hardware installation and maintenance, this work explores optimal data center resource management by balancing operational efficiency with fault tolerance through strategic rack positioning considering diverse resources and locations. Traditional mixed-integer programming (MIP) approaches often struggle with scalability, while heuristic methods may result in significant sub-optimality. To address these issues, this paper presents a novel two-tier optimization framework using a high-level deep reinforcement learning (DRL) model to guide a low-level gradient-based heuristic for local search. The high-level DRL agent employs Leader Reward for optimal rack type ordering, and the low-level heuristic efficiently maps racks to positions, minimizing movement counts and ensuring fault-tolerant resource distribution. This approach allows scalability to over 100,000 positions and 100 rack types. Our method outperformed the gradient-based heuristic by 7\% on average and the MIP solver by over 30\% in objective value. It achieved a 100\% success rate versus MIP's 97.5\% (within a 20-minute limit), completing in just 2 minutes compared to MIP's 1630 minutes (i.e., almost 4 orders of magnitude improvement). Unlike the MIP solver, which showed performance variability under time constraints and high penalties, our algorithm consistently delivered stable, efficient results - an essential feature for large-scale data center management.