Active Finite Reward Automaton Inference and Reinforcement Learning Using Queries and Counterexamples

The rapid advancement of artificial intelligence (AI) has accelerated the semiconductor industry, driving chip performance requirements to unprecedented levels. However, conventional system-on-chip (SoC) designs are increasingly constrained by the physical and economic limits of monolithic scaling, where continued transistor miniaturization yields diminishing returns and rising fabrication costs. To address these challenges, the industry has shifted toward chiplet-based heterogeneous integration and advanced packaging technologies. By interconnecting multiple specialized chiplets through high-density, high-bandwidth interfaces, this modular approach improves scalability, reusability, and cost efficiency while enabling faster time-to-market. However, chiplet-based systems introduce significant design complexity, particularly in signal integrity (SI) and power integrity (PI), where highly constrained interconnects and power delivery networks must meet stringent electrical requirements. Traditional design methodologies rely heavily on expert-driven manual tuning and extensive full-wave electromagnetic simulations using electronic design automation (EDA) tools, resulting in long design cycles that can span weeks or months. To overcome these limitations, this thesis investigates the application of AI, particularly deep reinforcement learning (RL), to automate and optimize chiplet-based system design with a focus on SI and PI challenges in advanced packaging technologies. The research initially explores a supervised learning approach using convolutional neural networks (CNNs) to classify signal impairments from eye diagrams. While effective, this approach reveals a critical dependence on large labeled datasets, which are scarce in SI/PI domains, motivating a shift toward RL for autonomous design space exploration without labeled data. The first major contribution proposes an RL framework for automated power distribution network (PDN) design in a four-chiplet system. Two RL algorithms, Dueling Double Deep Q-Network (DDDQN) and Proximal Policy Optimization (PPO), are evaluated under different reward strategies. Results show that dense analytical rewards based on power domain squareness and bump proximity enable the RL agent to achieve lower self-impedance than human-designed PDNs, ensuring power integrity across 28 power domains. The complete PDN layout, which typically requires three months of manual effort, is generated in under one hour using a single GPU. The second contribution focuses on signal integrity optimization for chiplet interconnects in CoWoS redistribution layers (RDLs). An RL agent is trained using a reward function that balances eye width, voltage transfer function (VTF) crosstalk, and interconnect density. The agent explores design variations across data rates from 4 to 32 GT/s using a simulation environment integrated with commercial EDA tools. Experimental results demonstrate up to 24.67% improvement in eye width and over 22 dB reduction in VTF crosstalk while satisfying UCIe electrical specifications. The training process also generates labeled datasets for future supervised learning. The third contribution introduces an RL-based routing framework for UCIe bridge interconnects in advanced System-in-Package (SiP). Using Proximal Policy Optimization, the framework learns to generate routing layouts that comply with constraints such as ground-signal-ground structures, layer assignment, and congestion avoidance. It supports all UCIe module configurations and produces design rule check-compliant solutions within 100 training episodes, reducing design time from four weeks to less than one hour while improving signal distribution and reducing via usage. In summary, this thesis presents a unified AI-driven methodology for addressing SI and PI challenges in chiplet-based systems. The proposed RL frameworks significantly outperform traditional design approaches in both efficiency and quality, demonstrating the potential of deep RL to enable scalable and intelligent design automation for advanced packaging and chiplet-based architectures.

A novel Deep Reinforcement Learning based automated stock trading system using cascaded LSTM networks

The application of reinforcement learning algorithms to motion planning is a research hotspot in robotics in recent years. However, training reinforcement learning agents from scratch has low training efficiency and difficulty in convergence. In this paper, a robot motion planning method based on residual reinforcement learning is proposed. This method divides the agent's policy of motion planning into initial policy and residual policy. The initial policy is composed of a neural network motion planner responsible for guiding the training direction of residual policy. The residual policy is composed of Proximal Policy Optimization (PPO) algorithm in reinforcement learning. A motion planning experiment is carried out in a simulation environment, and the result shows that the method can successfully perform motion planning. The comparison experiment between PPO and the proposed algorithm demonstrates that the proposed algorithm has better motion planning performance.

With the continuous progress of UAV technology and the rapid development of mobile edge computing (MEC), the UAV-assisted MEC system has shown great application potential in special fields such as disaster rescue and emergency response. However, traditional deep reinforcement learning (DRL) decision-making methods suffer from limitations such as difficulty in balancing multiple objectives and training convergence when making mixed action space decisions for UAV path planning and task offloading. This article innovatively proposes a hybrid decision framework based on the improved Dynamic Adaptive Genetic Optimization Algorithm (DAGOA) and soft actor–critic with hierarchical action decomposition, an uncertainty-quantified critic ensemble, and adaptive entropy temperature, where DAGOA performs an effective search and optimization in discrete action space, while SAC can perform fine control and adjustment in continuous action space. By combining the above algorithms, the joint optimization of drone path planning and task offloading can be achieved, improving the overall performance of the system. The experimental results show that the framework offers significant advantages in improving system performance, reducing energy consumption, and enhancing task completion efficiency. When the system adopts a hybrid decision framework, the reward score increases by a maximum of 153.53% compared to pure deep reinforcement learning algorithms for decision-making. Moreover, it can achieve an average improvement of 61.09% on the basis of various reinforcement learning algorithms such as proposed SAC, proximal policy optimization (PPO), deep deterministic policy gradient (DDPG), and twin delayed deep deterministic policy gradient (TD3).

In recent years, deep reinforcement learning (DRL) has made impressive achievements in many fields. However, existing DRL algorithms usually require a large amount of exploration to obtain a good action policy. In addition, in many complex situations, the reward function cannot be well designed to meet task requirements. These two problems will make it difficult for DRL to learn a good action policy within a relatively short period. The use of expert data can provide effective guidance and avoid unnecessary exploration. This study proposes a deep imitation reinforcement learning (DIRL) algorithm that uses a certain amount of expert demonstration data to speed up the training of DRL. In the proposed method, the learning agent imitates the expert's action policy by learning from demonstration data. After imitation learning, DRL is used to optimise the action policy in a self‐learning way. By experimental comparison on a video game called the Mario racing game, it is shown that the proposed DIRL algorithm with expert demonstration data can obtain much better performance than previous DRL algorithms without expert guidance.

Reinforcement learning, as a branch of machine learning, has been gradually applied in the control field. However, in the practical application of the algorithm, the hyperparametric approach to network settings for deep reinforcement learning still follows the empirical attempts of traditional machine learning (supervised learning and unsupervised learning). This method ignores part of the information generated by agents exploring the environment contained in the updating of the reinforcement learning value function, which will affect the performance of the convergence and cumulative return of reinforcement learning. The reinforcement learning algorithm based on dynamic parameter adjustment is a new method for setting learning rate parameters of deep reinforcement learning. Based on the traditional method of setting parameters for reinforcement learning, this method analyzes the advantages of different learning rates at different stages of reinforcement learning and dynamically adjusts the learning rates in combination with the temporal-difference (TD) error values to achieve the advantages of different learning rates in different stages to improve the rationality of the algorithm in practical application. At the same time, by combining the Robbins–Monro approximation algorithm and deep reinforcement learning algorithm, it is proved that the algorithm of dynamic regulation learning rate can theoretically meet the convergence requirements of the intelligent control algorithm. In the experiment, the effect of this method is analyzed through the continuous control scenario in the standard experimental environment of ”Car-on-The-Hill” of reinforcement learning, and it is verified that the new method can achieve better results than the traditional reinforcement learning in practical application. According to the model characteristics of the deep reinforcement learning, a more suitable setting method for the learning rate of the deep reinforcement learning network proposed. At the same time, the feasibility of the method has been proved both in theory and in the application. Therefore, the method of setting the learning rate parameter is worthy of further development and research.

A launch vehicle needs to adapt to a complex flight environment during flight, and traditional guidance and control algorithms can hardly deal with multi-factor uncertainties due to the high dependency on control models. To solve this problem, this paper designs a new intelligent flight control method for a rocket based on the deep reinforcement learning algorithm driven by knowledge and data. In this process, the Markov decision process of the rocket landing section is established by designing a reinforcement function with consideration of the combination effect on the return of the terminal constraint of the launch vehicle and the cumulative return of the flight process of the rocket. Meanwhile, to improve the training speed of the landing process of the launch vehicle and to enhance the generalization ability of the model, the strategic neural network model is obtained and trained via the form of a long short-term memory (LSTM) network combined with a full connection layer as a landing guidance strategy network. The proximal policy optimization (PPO) is the training algorithm of reinforcement learning network parameters combined with behavioral cloning (BC) as the reinforcement learning pre-training imitation learning algorithm. Notably, the rocket-borne environment is transplanted to the Nvidia Jetson TX2 embedded platform for the comparative testing and verification of this intelligent model, which is then used to generate real-time control commands for guiding the actual flying and landing process of the rocket. Further, comparisons of the results obtained from convex landing optimization and the proposed method in this work are performed to prove the effectiveness of this proposed method. The simulation results show that the intelligent control method in this work can meet the landing accuracy requirements of the launch vehicle with a fast convergence speed of 84 steps, and the decision time is only 2.5 ms. Additionally, it has the ability of online autonomous decision making as deployed on the embedded platform.

Energy-efficient and damage-recovery slithering gait design for a snake-like robot based on reinforcement learning and inverse reinforcement learning

Deep reinforcement learning is the technology of artificial neural networks in the field of decision-making and control. The traditional model-free reinforcement learning algorithm requires a large amount of environment interactive data to iterate the algorithm. This model’s performance also suffers due to low utilization of training data, while the model-based reinforcement learning (MBRL) algorithm improves the efficiency of the data, MBRL locks into low prediction accuracy. Although MBRL can utilize the additional data generated by the dynamic model, a system dynamics model with low prediction accuracy will provide low-quality data and affect the algorithm’s final result. In this paper, based on the A3C (Asynchronous Advantage Actor-Critic) algorithm, an improved model-based deep reinforcement learning algorithm using a learning degree network (MBRL-LDN) is presented. By comparing the differences between the predicted states outputted by the proposed multidynamic model and the original predicted states, the learning degree of the system dynamics model is calculated. The learning degree represents the quality of the data generated by the dynamic model and is used to decide whether to continue to interact with the dynamic model during a particular episode. Thus, low-quality data will be discarded. The superiority of the proposed method is verified by conducting extensive contrast experiments.

The flexible job shop scheduling problem (FJSP) holds significant importance in both theoretical research and practical applications. Given the complexity and diversity of FJSP, improving the generalization and quality of scheduling methods has become a hot topic of interest in both industry and academia. To address this, this paper proposes a Preference-Based Mask-PPO (PBMP) algorithm, which leverages the strengths of preference learning and invalid action masking to optimize FJSP solutions. First, a reward predictor based on preference learning is designed to model reward prediction by comparing random fragments, eliminating the need for complex reward function design. Second, a novel intelligent switching mechanism is introduced, where proximal policy optimization (PPO) is employed to enhance exploration during sampling, and masked proximal policy optimization (Mask-PPO) refines the action space during training, significantly improving efficiency and solution quality. Furthermore, the Pearson correlation coefficient (PCC) is used to evaluate the performance of the preference model. Finally, comparative experiments on FJSP benchmark instances of varying sizes demonstrate that PBMP outperforms traditional scheduling strategies such as dispatching rules, OR-Tools, and other deep reinforcement learning (DRL) algorithms, achieving superior scheduling policies and faster convergence. Even with increasing instance sizes, preference learning proves to be an effective reward mechanism in reinforcement learning for FJSP. The ablation study further highlights the advantages of each key component in the PBMP algorithm across performance metrics.

Attempts to address the production scheduling problem thus far rely on simplifying assumptions, such as static environment and inflexible size of the problem, which compromises the schedule performance in practice due to many unpredictable disruptions to the system. Thus, the study of scheduling in the presence of real-time events, termed dynamic scheduling, continues to attract attention given the agility, flexibility, and timeliness modern production systems must deliver. Additionally, the changing nature of the manufacturing system also raises new challenges to existing scheduling strategies. At the front-end, the development of advanced data creation and exchange frameworks such as the Internet of things and cyber-physical system and their applications to the industrial environment have created an abundance of industrial data, while at the backend, edge and cloud computing technologies greatly enhance the capacity to process that data. Industrial data must be mined and analyzed so that the investment in infrastructure is not wasted, and the production system managed more effectively and in real-time. Many data-driven technologies have been adopted in scheduling research, a promising candidate among them being reinforcement learning (RL) which is able to build a direct mapping from observation of environment to actions that improve its performance. In this thesis, a deep multi-agent reinforcement learning (deep MARL) architecture is proposed to solve the dynamic scheduling problem (DSP). The deep reinforcement learning (DRL) algorithm is used to train the decentralized scheduling agents, to capture the relationship between information on the factory floor and scheduling objectives, with the aim of making real-time decisions for a manufacturing system with frequent unexpected events. Two major aspects of deep MARL application to DSP are addressed in this work, namely the conversion from traditional static scheduling problem (SSP) to dynamic scheduling in a practical context, and the adaptation of existing deep MARL algorithms to solve the scheduling problem in such an environment. Some impractical constraints of traditional studies are removed to create a research context that is closer to actual practice, result in a scheduling problem of variable size and scope. Specialized state and action representations that can handle the ever-changing specification of problem are developed; the criteria of feature selection in dynamic environment are also discussed. Recent progressions in DRL and MARL research are integrated into the proposed approach after selection and adaptation. In addition, various improvements to common deep MARL architecture are proposed, including the lightweight multilayer perceptron (MLP) encoder that is efficient in handling unstructured industrial data, a training scheme under the multi-agent architecture to improve the stability of training and overall performance, and knowledge-based reward-shaping techniques to decompose the joint reward signal into individual utilities to speed up the learning and encourage cooperative behavior between agents. Simulation studies are then conducted for the ablation study and validation. In the first stage, the performance of the proposed approach, either as individual components or as an integrated model, are tested in iterative simulation runs within which a unique instance of production is created. Meanwhile, a set of DRL-based approaches from recent publications are run in parallel. Results suggest that the contribution of each improvement is significant; the integrated architecture also delivers stronger performance than peer DRL-based approaches. For the validation, a set of priority rules that have strong performance in specified context and are widely applied in actual production scheduling are used as the benchmark. Proposed approach also provides performance gain compared to the strongest rule, with a minor increase in computation cost and negligible latency in decision-making.

Reinforcement learning is a mathematical framework for agents to interact intelligently with their environment. Unlike supervised learning, where a system learns with the help of labeled data, reinforcement learning agents learn how to act by trial and error only receiving a reward signal from their environments. A field where reinforcement learning has been prominently successful is robotics [3]. However, real-world control problems are also particularly challenging because of the noise and high- dimensionality of input data (e.g., visual input). In recent years, in the field of supervised learning, deep neural networks have been successfully used to extract meaning from this kind of data. Building on these advances, deep reinforcement learning was used to solve complex problems like Atari games and Go. Mnih et al. [1] built a system with fixed hyper parameters able to learn to play 49 different Atari games only from raw pixel inputs. However, in order to apply the same methods to real-world control problems, deep reinforcement learning has to be able to deal with continuous action spaces. Discretizing continuous action spaces would scale poorly, since the number of discrete actions grows exponentially with the dimensionality of the action. Furthermore, having a parametrized policy can be advantageous because it can generalize in the action space. Therefore with this thesis we study state-of-the-art deep reinforcement learning algorithm, Deep Deterministic Policy Gradients. We provide a theoretical comparison to other popular methods, an evaluation of its performance, identify its limitations and investigate future directions of research. The remainder of the thesis is organized as follows. We start by introducing the field of interest, machine learning, focusing our attention of deep learning and reinforcement learning. We continue by describing in details the two main algorithms, core of this study, namely Deep Q-Network (DQN) and Deep Deterministic Policy Gradients (DDPG). We then provide implementatory details of DDPG and our test environment, followed by a description of benchmark test cases. Finally, we discuss the results of our evaluation, identifying limitations of the current approach and proposing future avenues of research.

The transition to sustainable and intelligent energy systems has intensified the development of smart microgrids, which offer decentralized, resilient, and efficient power solutions. This review critically examines the integration of Artificial Intelligence (AI) and Deep Reinforcement Learning (DRL) into smart microgrid platforms, focusing on their role in optimizing sustainable energy management. Traditional energy management systems often struggle to adapt to the dynamic nature of modern energy demands, renewable energy intermittency, and grid complexity. AI-driven solutions, particularly DRL, provide adaptive, autonomous, and data-driven mechanisms for real-time decision-making and predictive control within microgrids. DRL, by learning optimal policies through interaction with the environment, is capable of handling multi-objective problems, including demand-response optimization, energy storage control, load forecasting, and distributed generation scheduling. This paper synthesizes recent advancements and applications of DRL algorithms such as Deep Q-Networks (DQN), Deep Deterministic Policy Gradient (DDPG), and Proximal Policy Optimization (PPO) in smart microgrids. It also explores hybrid models that combine DRL with other AI techniques, such as fuzzy logic and neural networks, to improve performance under uncertainty and nonlinearity. Furthermore, the review evaluates benchmark testbeds, simulation tools, and real-time platforms used to implement and validate these intelligent systems. Challenges such as high computational costs, model generalizability, real-time implementation issues, and cybersecurity vulnerabilities are discussed. The study also highlights recent research trends emphasizing decentralized control, edge computing, and federated learning to enhance scalability and privacy in DRL-based microgrid applications. Future directions suggest the need for explainable AI (XAI), robust training environments, and standardization to facilitate the wider adoption of DRL in real-world microgrid systems. The integration of DRL into smart microgrids represents a transformative shift toward resilient, efficient, and sustainable energy ecosystems. This review offers a comprehensive understanding of how AI and DRL are revolutionizing energy management, providing a foundation for future research and practical deployment in smart energy infrastructures.

In this article, we consider a subclass of partially observable Markov decision process (POMDP) problems which we termed confounding POMDPs. In these types of POMDPs, temporal difference (TD)-based reinforcement learning (RL) algorithms struggle, as TD error cannot be easily derived from observations. We solve these types of problems using a new bio-inspired neural architecture that combines a modulated Hebbian network (MOHN) with deep Q-network (DQN), which we call modulated Hebbian plus Q-network architecture (MOHQA). The key idea is to use a Hebbian network with rarely correlated bio-inspired neural traces to bridge temporal delays between actions and rewards when confounding observations and sparse rewards result in inaccurate TD errors. In MOHQA, DQN learns low-level features and control, while the MOHN contributes to high-level decisions by associating rewards with past states and actions. Thus, the proposed architecture combines two modules with significantly different learning algorithms, a Hebbian associative network and a classical DQN pipeline, exploiting the advantages of both. Simulations on a set of POMDPs and on the Malmo environment show that the proposed algorithm improved DQN's results and even outperformed control tests with advantage-actor critic (A2C), quantile regression DQN with long short-term memory (QRDQN + LSTM), Monte Carlo policy gradient (REINFORCE), and aggregated memory for reinforcement learning (AMRL) algorithms on most difficult POMDPs with confounding stimuli and sparse rewards.

Learning ability is the basic characteristic of human intelligence. The July 1, 2005 issue of Science published a list of 125 important questions in science. Among them, the question 94 “What are the limits of learning by machines?”. The annotation “Computers can already beat the world’s best chess players, and they have a wealth of information on the Web to draw on. But abstract reasoning is still beyond any machine”. In recent artificial intelligence has made great progresses. In 1997, the rise of the man-machine war, IBM Supercomputer Deep Blue defeated the chess master Garry Kasparov. On February 14, 2011, IBM’s Watson supercomputer won a practice round against Jeopardy champions Ken Jennings and Brad Rutter. In March 2016, Google DeepMind’s AlphaGo sealed a 4-1 victory over a South Korean Go grandmaster Lee Se-dol. This paper focuses on the machine learning methods of AlphaGo, including reinforcement learning, deep learning, deep reinforcement learning, analysis of the existing problems and the latest research progress. Deep reinforcement learning is the combination of deep learning and reinforcement learning, which can realize the learning algorithm from the perception to action. Simply said, this is the same as human behavior, input sensing information such as vision, and then, direct output action through the deep neural network. Deep reinforcement learning has the potential to learn a variety of skills for the robot to achieve full autonomy. Even though reinforcement learning is practiced successfully, but feature states need to manually set, for complex scene is a difficult thing, especially easy to cause the dimension disaster, and expression is not good. In 2010, Sascha Lange and Martin Riedmiller proposed deep auto-encoder neural networks in reinforcement learning to extract feature, which is used to control the visual correlation. In 2013, DeepMind proposed deep Q-network (DQN) in NIPS 2013, using convolution neural network to extract features, and then applied in reinforcement learning. They continue to improve and published an improved version of DQN on Nature in 2015, which has aroused widespread concern. In order to break through the limits of learning by machines, cognitive machine learning is proposed, which is the combination of machine learning and brain cognition, so that the machine intelligence is constantly evolving, and gradually reaches the human level of artificial intelligence. A cognitive model entitled Consciousness And Memory (CAM) is proposed by author, which consists of memory, consciousness, high-level cognitive functions, perception and motor. High-level cognitive functions of the brain include learning, language, thinking, decision making, emotion, and so on. Learning is a course to accept the stimulus through the nervous system and obtain new behavior, habits and accumulation experience. According to the current research progress of brain science and cognitive science, cognitive machine learning may be interested in learning emergence, procedural memory knowledge learning, learning evolution and so on. For intelligence, so-called evolution is refers to the learning of learning and the structure also follows the change. It is important to record the learning result by structure changing and improve the learning method.