As the most important auxiliary transportation equipment in coal mines, mining electric locomotives are mostly operated manually at present. However, due to the complex and ever-changing coal mine environment, electric locomotive safety accidents occur frequently these years. A mining electric locomotive control method that can adapt to different complex mining environments is needed. Reinforcement Learning (RL) is concerned with how artificial agents ought to take actions in an environment so as to maximize reward, which can help achieve automatic control of mining electric locomotive. In this paper, we present how to apply RL to the autonomous control of mining electric locomotives. To achieve more precise control, we further propose an improved epsilon-greedy (IEG) algorithm which can better balance the exploration and exploitation. To verify the effectiveness of this method, a co-simulation platform for autonomous control of mining electric locomotives is built which can complete closed-loop simulation of the vehicles. The simulation results show that this method ensures the locomotives following the front vehicle safely and responding promptly in the event of sudden obstacles on the road when the vehicle in complex and uncertain coal mine environments.
The development of intelligent robots requires control policies that can handle dynamic environments and evolving tasks. Pre-training reinforcement learning has emerged as an effective approach to address these demands by enabling robots to acquire reusable motor skills. However, they often rely on large datasets or expert-designed goal spaces, limiting adaptability. Additionally, these methods need help to generate dynamic and diverse skills in high-dimensional state spaces, reducing their effectiveness for downstream tasks. In this paper, we propose CMS-PRL, a pre-training reinforcement learning method inspired by the Central Motor System (CMS). First, we introduce a fusion reward mechanism that combines the basic motor reward with mutual information reward, promoting the discovery of dynamic skills during pre-training without reliance on external data. Second, we design a skill encoding method inspired by the motor program of the basal ganglia, providing rich and continuous skill instructions during pre-training. Finally, we propose a skill activity function to regulate motor skill activity, enabling the generation of skills with different activity levels, thereby enhancing the robot's flexibility in downstream tasks. We evaluate the model on four types of robots in a challenging set of sparse-reward tasks. Experimental results demonstrate that CMS-PRL generates diverse, reusable motor skills to solve various downstream tasks and outperforms baseline methods, particularly in high-degree-of-freedom robots and complex tasks.
This study addresses a gap in the utilization of Reinforcement Learning (RL) and Machine Learning (ML) techniques in solving the Stochastic Vehicle Routing Problem (SVRP) that involves the challenging task of optimizing vehicle routes under uncertain conditions. We propose a novel end-to-end framework that comprehensively addresses the key sources of stochasticity in SVRP and utilizes an RL agent with a simple yet effective architecture and a tailored training method. Through comparative analysis, our proposed model demonstrates superior performance compared to a widely adopted state-of-the-art metaheuristic, achieving a significant 3.43% reduction in travel costs. Furthermore, the model exhibits robustness across diverse SVRP settings, highlighting its adaptability and ability to learn optimal routing strategies in varying environments. The publicly available implementation of our framework serves as a valuable resource for future research endeavors aimed at advancing RL-based solutions for SVRP.
Motivated by policy gradient methods in the context of reinforcement learning, we identify a large deviation rate function for the iterates generated by stochastic gradient descent for possibly non-convex objectives satisfying a Polyak-{\L}ojasiewicz condition. Leveraging the contraction principle from large deviations theory, we illustrate the potential of this result by showing how convergence properties of policy gradient with a softmax parametrization and an entropy regularized objective can be naturally extended to a wide spectrum of other policy parametrizations.
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.
Quadruped robots have emerged as an evolving technology that currently leverages simulators to develop a robust controller capable of functioning in the real-world without the need for further training. However, since it is impossible to predict all possible real-world situations, our research explores the possibility of enabling them to continue learning even after their deployment. To this end, we designed two continual learning scenarios, sequentially training the robot on different environments while simultaneously evaluating its performance across all of them. Our approach sheds light on the extent of both forward and backward skill transfer, as well as the degree to which the robot might forget previously acquired skills. By addressing these factors, we hope to enhance the adaptability and performance of quadruped robots in real-world scenarios.
Tactile information is important for robust performance in robotic tasks that involve physical interaction, such as object manipulation. However, with more data included in the reasoning and control process, modeling behavior becomes increasingly difficult. Deep Reinforcement Learning (DRL) produced promising results for learning complex behavior in various domains, including tactile-based manipulation in robotics. In this work, we present our open-source reinforcement learning environments for the TIAGo service robot. They produce tactile sensor measurements that resemble those of a real sensorised gripper for TIAGo, encouraging research in transfer learning of DRL policies. Lastly, we show preliminary training results of a learned force control policy and compare it to a classical PI controller.
Several recent works have proposed instance-dependent upper bounds on the number of episodes needed to identify, with probability $1-\delta$, an $\varepsilon$-optimal policy in finite-horizon tabular Markov Decision Processes (MDPs). These upper bounds feature various complexity measures for the MDP, which are defined based on different notions of sub-optimality gaps. However, as of now, no lower bound has been established to assess the optimality of any of these complexity measures, except for the special case of MDPs with deterministic transitions. In this paper, we propose the first instance-dependent lower bound on the sample complexity required for the PAC identification of a near-optimal policy in any tabular episodic MDP. Additionally, we demonstrate that the sample complexity of the PEDEL algorithm of \cite{Wagenmaker22linearMDP} closely approaches this lower bound. Considering the intractability of PEDEL, we formulate an open question regarding the possibility of achieving our lower bound using a computationally-efficient algorithm.
Reinforcement learning is an emerging approach to control dynamical systems for which classical approaches are difficult to apply. However, trained agents may not generalize against the variations of system parameters. This paper presents the concept of dual-scale homogeneity, an important property in understating the scaling behavior of nonlinear systems. Furthermore, it also presents an effective yet simple approach to designing a parameter-dependent control law that homogenizes a nonlinear system. The presented approach is applied to two systems, demonstrating its ability to provide a consistent performance irrespective of parameters variations. To demonstrate the practicality of the proposed approach, the control policy is generated by a deep deterministic policy gradient to control the load position of a quadrotor with a slung load. The proposed synergy between the homogeneity transformations and reinforcement learning yields superior performance compared to other recent learning-based control techniques. It achieves a success rate of 96% in bringing the load to its designated target with a 3D RMSE of 0.0253 m. The video that shows the experimental results along with a summary of the paper is available at this link.
Multi-Agent Reinforcement Learning is becoming increasingly more important in times of autonomous driving and other smart industrial applications. Simultaneously a promising new approach to Reinforcement Learning arises using the inherent properties of quantum mechanics, reducing the trainable parameters of a model significantly. However, gradient-based Multi-Agent Quantum Reinforcement Learning methods often have to struggle with barren plateaus, holding them back from matching the performance of classical approaches. While gradient free Quantum Reinforcement Learning methods may alleviate some of these challenges, they too are not immune to the difficulties posed by barren plateaus. We build upon an existing approach for gradient free Quantum Reinforcement Learning and propose three genetic variations with Variational Quantum Circuits for Multi-Agent Reinforcement Learning using evolutionary optimization. We evaluate our genetic variations in the Coin Game environment and also compare them to classical approaches. We showed that our Variational Quantum Circuit approaches perform significantly better compared to a neural network with a similar amount of trainable parameters. Compared to the larger neural network, our approaches archive similar results using $97.88\%$ less parameters.
We introduce and study constrained Markov Decision Processes (cMDPs) with anytime constraints. An anytime constraint requires the agent to never violate its budget at any point in time, almost surely. Although Markovian policies are no longer sufficient, we show that there exist optimal deterministic policies augmented with cumulative costs. In fact, we present a fixed-parameter tractable reduction from anytime-constrained cMDPs to unconstrained MDPs. Our reduction yields planning and learning algorithms that are time and sample-efficient for tabular cMDPs so long as the precision of the costs is logarithmic in the size of the cMDP. However, we also show that computing non-trivial approximately optimal policies is NP-hard in general. To circumvent this bottleneck, we design provable approximation algorithms that efficiently compute or learn an arbitrarily accurate approximately feasible policy with optimal value so long as the maximum supported cost is bounded by a polynomial in the cMDP or the absolute budget. Given our hardness results, our approximation guarantees are the best possible under worst-case analysis.
In this study, we have developed a dynamic asset allocation investment strategy using reinforcement learning techniques. To begin with, we have addressed the crucial issue of incorporating non-stationarity of financial time series data into reinforcement learning algorithms, which is a significant implementation in the application of reinforcement learning in investment strategies. Our findings highlight the significance of introducing certain variables such as regime change in the environment setting to enhance the prediction accuracy. Furthermore, the application of reinforcement learning in investment strategies provides a remarkable advantage of setting the optimization problem flexibly. This enables the integration of practical constraints faced by investors into the algorithm, resulting in efficient optimization. Our study has categorized the investment strategy formulation conditions into three main categories, including performance measurement indicators, portfolio management rules, and other constraints. We have evaluated the impact of incorporating these conditions into the environment and rewards in a reinforcement learning framework and examined how they influence investment behavior.
We introduce a biologically plausible RL framework for solving tasks in partially observable Markov decision processes (POMDPs). The proposed algorithm combines three integral parts: (1) A Meta-RL architecture, resembling the mammalian basal ganglia; (2) A biologically plausible reinforcement learning algorithm, exploiting temporal difference learning and eligibility traces to train the policy and the value-function; (3) An online automatic differentiation algorithm for computing the gradients with respect to parameters of a shared recurrent network backbone. Our experimental results show that the method is capable of solving a diverse set of partially observable reinforcement learning tasks. The algorithm we call real-time recurrent reinforcement learning (RTRRL) serves as a model of learning in biological neural networks, mimicking reward pathways in the basal ganglia.
Hamiltonian simulation is believed to be one of the first tasks where quantum computers can yield a quantum advantage. One of the most popular methods of Hamiltonian simulation is Trotterization, which makes use of the approximation $e^{i\sum_jA_j}\sim \prod_je^{iA_j}$ and higher-order corrections thereto. However, this leaves open the question of the order of operations (i.e. the order of the product over $j$, which is known to affect the quality of approximation). In some cases this order is fixed by the desire to minimise the error of approximation; when it is not the case, we propose that the order can be chosen to optimize compilation to a native quantum architecture. This presents a new compilation problem -- order-agnostic quantum circuit compilation -- which we prove is NP-hard in the worst case. In lieu of an easily-computable exact solution, we turn to methods of heuristic optimization of compilation. We focus on reinforcement learning due to the sequential nature of the compilation task, comparing it to simulated annealing and Monte Carlo tree search. While two of the methods outperform a naive heuristic, reinforcement learning clearly outperforms all others, with a gain of around 12% with respect to the second-best method and of around 50% compared to the naive heuristic in terms of the gate count. We further test the ability of RL to generalize across instances of the compilation problem, and find that a single learner is able to solve entire problem families. This demonstrates the ability of machine learning techniques to provide assistance in an order-agnostic quantum compilation task.
To promote cooperation and strengthen the individual impact on the collective outcome in social dilemmas, we propose the Environmental-impact Multi-Agent Reinforcement Learning (EMuReL) method where each agent estimates the "environmental impact" of every other agent, that is, the difference in the current environment state compared to the hypothetical environment in the absence of that other agent. Inspired by the Inequity Aversion model, the agent then compares its own reward with those of its fellows multiplied by their environmental impacts. If its reward exceeds the scaled reward of one of its fellows, the agent takes "social responsibility" toward that fellow by reducing its own reward. Therefore, the less influential an agent is in reaching the current state, the more social responsibility is taken by other agents. Experiments in the Cleanup (resp. Harvest) test environment demonstrate that agents trained based on EMuReL learn to cooperate more effectively and obtain $54\%$ ($39\%$) and $20\%$ ($44\%$) more total rewards while preserving the same cooperation levels compared to when they are trained based on the two state-of-the-art reward reshaping methods inequity aversion and social influence.
In human societies, people often incorporate fairness in their decisions and treat reciprocally by being kind to those who act kindly. They evaluate the kindness of others' actions not only by monitoring the outcomes but also by considering the intentions. This behavioral concept can be adapted to train cooperative agents in Multi-Agent Reinforcement Learning (MARL). We propose the KindMARL method, where agents' intentions are measured by counterfactual reasoning over the environmental impact of the actions that were available to the agents. More specifically, the current environment state is compared with the estimation of the current environment state provided that the agent had chosen another action. The difference between each agent's reward, as the outcome of its action, with that of its fellow, multiplied by the intention of the fellow is then taken as the fellow's "kindness". If the result of each reward-comparison confirms the agent's superiority, it perceives the fellow's kindness and reduces its own reward. Experimental results in the Cleanup and Harvest environments show that training based on the KindMARL method enabled the agents to earn 89\% (resp. 37\%) and 44% (resp. 43\%) more total rewards than training based on the Inequity Aversion and Social Influence methods. The effectiveness of KindMARL is further supported by experiments in a traffic light control problem.
Stateful policies play an important role in reinforcement learning, such as handling partially observable environments, enhancing robustness, or imposing an inductive bias directly into the policy structure. The conventional method for training stateful policies is Backpropagation Through Time (BPTT), which comes with significant drawbacks, such as slow training due to sequential gradient propagation and the occurrence of vanishing or exploding gradients. The gradient is often truncated to address these issues, resulting in a biased policy update. We present a novel approach for training stateful policies by decomposing the latter into a stochastic internal state kernel and a stateless policy, jointly optimized by following the stateful policy gradient. We introduce different versions of the stateful policy gradient theorem, enabling us to easily instantiate stateful variants of popular reinforcement learning and imitation learning algorithms. Furthermore, we provide a theoretical analysis of our new gradient estimator and compare it with BPTT. We evaluate our approach on complex continuous control tasks, e.g., humanoid locomotion, and demonstrate that our gradient estimator scales effectively with task complexity while offering a faster and simpler alternative to BPTT.
Many capable large language models (LLMs) are developed via self-supervised pre-training followed by a reinforcement-learning fine-tuning phase, often based on human or AI feedback. During this stage, models may be guided by their inductive biases to rely on simpler features which may be easier to extract, at a cost to robustness and generalisation. We investigate whether principles governing inductive biases in the supervised fine-tuning of LLMs also apply when the fine-tuning process uses reinforcement learning. Following Lovering et al (2021), we test two hypotheses: that features more $\textit{extractable}$ after pre-training are more likely to be utilised by the final policy, and that the evidence for/against a feature predicts whether it will be utilised. Through controlled experiments on synthetic and natural language tasks, we find statistically significant correlations which constitute strong evidence for these hypotheses.
Digital twins promise to revolutionize engineering by offering new avenues for optimization, control, and predictive maintenance. We propose a novel framework for simultaneously training the digital twin of an engineering system and an associated control agent. The twin's training combines adjoint-based data assimilation and system identification methods, while the control agent's training merges model-based optimal control with model-free reinforcement learning. The control agent evolves along two independent paths: one driven by model-based optimal control and the other by reinforcement learning. The digital twin serves as a virtual environment for confrontation and indirect interaction, functioning as an "expert demonstrator." The best policy is selected for real-world interaction and cloned to the other path if training stagnates. We call this framework Reinforcement Twinning (RT). The framework is tested on three diverse engineering systems and control tasks: (1) controlling a wind turbine under varying wind speeds, (2) trajectory control of flapping-wing micro air vehicles (FWMAVs) facing wind gusts, and (3) mitigating thermal loads in managing cryogenic storage tanks. These test cases use simplified models with known ground truth closure laws. Results show that the adjoint-based digital twin training is highly sample-efficient, completing within a few iterations. For the control agent training, both model-based and model-free approaches benefit from their complementary learning experiences. The promising results pave the way for implementing the RT framework on real systems.
Meta Reinforcement Learning (Meta RL) trains agents that adapt to fast-changing environments and tasks. Current strategies often lose adaption efficiency due to the passive nature of model exploration, causing delayed understanding of new transition dynamics. This results in particularly fast-evolving tasks being impossible to solve. We propose a novel approach, Hypothesis Network Planned Exploration (HyPE), that integrates an active and planned exploration process via the hypothesis network to optimize adaptation speed. HyPE uses a generative hypothesis network to form potential models of state transition dynamics, then eliminates incorrect models through strategically devised experiments. Evaluated on a symbolic version of the Alchemy game, HyPE outpaces baseline methods in adaptation speed and model accuracy, validating its potential in enhancing reinforcement learning adaptation in rapidly evolving settings.