In urban planning, land use readjustment plays a pivotal role in aligning land use configurations with the current demands for sustainable urban development. However, present-day urban planning practices face two main issues. Firstly, land use decisions are predominantly dependent on human experts. Besides, while resident engagement in urban planning can promote urban sustainability and livability, it is challenging to reconcile the diverse interests of stakeholders. To address these challenges, we introduce a Consensus-based Multi-Agent Reinforcement Learning framework for real-world land use readjustment. This framework serves participatory urban planning, allowing diverse intelligent agents as stakeholder representatives to vote for preferred land use types. Within this framework, we propose a novel consensus mechanism in reward design to optimize land utilization through collective decision making. To abstract the structure of the complex urban system, the geographic information of cities is transformed into a spatial graph structure and then processed by graph neural networks. Comprehensive experiments on both traditional top-down planning and participatory planning methods from real-world communities indicate that our computational framework enhances global benefits and accommodates diverse interests, leading to improved satisfaction across different demographic groups. By integrating Multi-Agent Reinforcement Learning, our framework ensures that participatory urban planning decisions are more dynamic and adaptive to evolving community needs and provides a robust platform for automating complex real-world urban planning processes.
The technique of Reinforcement Learning from Human Feedback (RLHF) is a commonly employed method to improve pre-trained Language Models (LM), enhancing their ability to conform to human preferences. Nevertheless, the current RLHF-based LMs necessitate full retraining each time novel queries or feedback are introduced, which becomes a challenging task because human preferences can vary between different domains or tasks. Retraining LMs poses practical difficulties in many real-world situations due to the significant time and computational resources required, along with concerns related to data privacy. To address this limitation, we propose a new method called Continual Optimal Policy Regularization (COPR), in which we compute the distribution of optimal policy bypassing the partition function and then regularize the current policy based on the historically optimal distribution to mitigate Catastrophic Forgetting (CF). COPR involves a single learning phase and doesn't necessitate complex reinforcement learning. Importantly, it shares the capability with RLHF to learn from unlabeled data by maintaining a scoring module, similar to reward model, making it flexible for continually learning without human feedback. Our experimental results show that COPR outperforms strong Continuous Learning (CL) baselines when it comes to consistently aligning with human preferences on incremental tasks and domains.
We propose a new regret minimization algorithm for episodic sparse linear Markov decision process (SMDP) where the state-transition distribution is a linear function of observed features. The only previously known algorithm for SMDP requires the knowledge of the sparsity parameter and oracle access to an unknown policy. We overcome these limitations by combining the doubly robust method that allows one to use feature vectors of \emph{all} actions with a novel analysis technique that enables the algorithm to use data from all periods in all episodes. The regret of the proposed algorithm is $\tilde{O}(\sigma^{-1}_{\min} s_{\star} H \sqrt{N})$, where $\sigma_{\min}$ denotes the restrictive the minimum eigenvalue of the average Gram matrix of feature vectors, $s_\star$ is the sparsity parameter, $H$ is the length of an episode, and $N$ is the number of rounds. We provide a lower regret bound that matches the upper bound up to logarithmic factors on a newly identified subclass of SMDPs. Our numerical experiments support our theoretical results and demonstrate the superior performance of our algorithm.
Reinforcement learning from human feedback (RLHF) enables machine learning systems to learn objectives from human feedback. A core limitation of these systems is their assumption that all feedback comes from a single human teacher, despite querying a range of distinct teachers. We propose the Hidden Utility Bandit (HUB) framework to model differences in teacher rationality, expertise, and costliness, formalizing the problem of learning from multiple teachers. We develop a variety of solution algorithms and apply them to two real-world domains: paper recommendation systems and COVID-19 vaccine testing. We find that the Active Teacher Selection (ATS) algorithm outperforms baseline algorithms by actively selecting when and which teacher to query. The HUB framework and ATS algorithm demonstrate the importance of leveraging differences between teachers to learn accurate reward models, facilitating future research on active teacher selection for robust reward modeling.
The boundary control problem is a non-convex optimization and control problem in many scientific domains, including fluid mechanics, structural engineering, and heat transfer optimization. The aim is to find the optimal values for the domain boundaries such that the enclosed domain adhering to the governing equations attains the desired state values. Traditionally, non-linear optimization methods, such as the Interior-Point method (IPM), are used to solve such problems. This project explores the possibilities of using deep learning and reinforcement learning to solve boundary control problems. We adhere to the framework of iterative optimization strategies, employing a spatial neural network to construct well-informed initial guesses, and a spatio-temporal neural network learns the iterative optimization algorithm using policy gradients. Synthetic data, generated from the problems formulated in the literature, is used for training, testing and validation. The numerical experiments indicate that the proposed method can rival the speed and accuracy of existing solvers. In our preliminary results, the network attains costs lower than IPOPT, a state-of-the-art non-linear IPM, in 51\% cases. The overall number of floating point operations in the proposed method is similar to that of IPOPT. Additionally, the informed initial guess method and the learned momentum-like behaviour in the optimizer method are incorporated to avoid convergence to local minima.
In this paper we investigate transformer architectures designed for partially observable online reinforcement learning. The self-attention mechanism in the transformer architecture is capable of capturing long-range dependencies and it is the main reason behind its effectiveness in processing sequential data. Nevertheless, despite their success, transformers have two significant drawbacks that still limit their applicability in online reinforcement learning: (1) in order to remember all past information, the self-attention mechanism requires access to the whole history to be provided as context. (2) The inference cost in transformers is expensive. In this paper, we introduce recurrent alternatives to the transformer self-attention mechanism that offer context-independent inference cost, leverage long-range dependencies effectively, and performs well in online reinforcement learning task. We quantify the impact of the different components of our architecture in a diagnostic environment and assess performance gains in 2D and 3D pixel-based partially-observable environments (e.g. T-Maze, Mystery Path, Craftax, and Memory Maze). Compared with a state-of-the-art architecture, GTrXL, inference in our approach is at least 40% cheaper while reducing memory use more than 50%. Our approach either performs similarly or better than GTrXL, improving more than 37% upon GTrXL performance in harder tasks.
Vessel navigation is difficult in restricted waterways and in the presence of static and dynamic obstacles. This difficulty can be attributed to the high-level decisions taken by humans during these maneuvers, which is evident from the fact that 85% of the reported marine accidents are traced back to human errors. Artificial intelligence-based methods offer us a way to eliminate human intervention in vessel navigation. Newer methods like Deep Reinforcement Learning (DRL) can optimize multiple objectives like path following and collision avoidance at the same time while being computationally cheaper to implement in comparison to traditional approaches. Before addressing the challenge of collision avoidance along with path following, the performance of DRL-based controllers on the path following task alone must be established. Therefore, this study trains a DRL agent using Proximal Policy Optimization (PPO) algorithm and tests it against a traditional PD controller guided by an Integral Line of Sight (ILOS) guidance system. The Krisco Container Ship (KCS) is chosen to test the different controllers. The ship dynamics are mathematically simulated using the Maneuvering Modelling Group (MMG) model developed by the Japanese. The simulation environment is used to train the deep reinforcement learning-based controller and is also used to tune the gains of the traditional PD controller. The effectiveness of the controllers in the presence of wind is also investigated.
Human decision-making errors cause a majority of globally reported marine accidents. As a result, automation in the marine industry has been gaining more attention in recent years. Obstacle avoidance becomes very challenging for an autonomous surface vehicle in an unknown environment. We explore the feasibility of using Deep Q-Learning (DQN), a deep reinforcement learning approach, for controlling an underactuated autonomous surface vehicle to follow a known path while avoiding collisions with static and dynamic obstacles. The ship's motion is described using a three-degree-of-freedom (3-DOF) dynamic model. The KRISO container ship (KCS) is chosen for this study because it is a benchmark hull used in several studies, and its hydrodynamic coefficients are readily available for numerical modelling. This study shows that Deep Reinforcement Learning (DRL) can achieve path following and collision avoidance successfully and can be a potential candidate that may be investigated further to achieve human-level or even better decision-making for autonomous marine vehicles.
Traditional controllers have limitations as they rely on prior knowledge about the physics of the problem, require modeling of dynamics, and struggle to adapt to abnormal situations. Deep reinforcement learning has the potential to address these problems by learning optimal control policies through exploration in an environment. For safety-critical environments, it is impractical to explore randomly, and replacing conventional controllers with black-box models is also undesirable. Also, it is expensive in continuous state and action spaces, unless the search space is constrained. To address these challenges we propose a specialized deep residual policy safe reinforcement learning with a cycle of learning approach adapted for complex and continuous state-action spaces. Residual policy learning allows learning a hybrid control architecture where the reinforcement learning agent acts in synchronous collaboration with the conventional controller. The cycle of learning initiates the policy through the expert trajectory and guides the exploration around it. Further, the specialization through the input-output hidden Markov model helps to optimize policy that lies within the region of interest (such as abnormality), where the reinforcement learning agent is required and is activated. The proposed solution is validated on the Tennessee Eastman process control.
In Reinforcement Learning (RL), agents aim at maximizing cumulative rewards in a given environment. During the learning process, RL agents face the dilemma of exploitation and exploration: leveraging existing knowledge to acquire rewards or seeking potentially higher ones. Using uncertainty as a guiding principle provides an active and effective approach to solving this dilemma and ensemble-based methods are one of the prominent avenues for quantifying uncertainty. Nevertheless, conventional ensemble-based uncertainty estimation lacks an explicit prior, deviating from Bayesian principles. Besides, this method requires diversity among members to generate less biased uncertainty estimation results. To address the above problems, previous research has incorporated random functions as priors. Building upon these foundational efforts, our work introduces an innovative approach with delicately designed prior NNs, which can incorporate maximal diversity in the initial value functions of RL. Our method has demonstrated superior performance compared with the random prior approaches in solving classic control problems and general exploration tasks, significantly improving sample efficiency.
Energy storage systems (ESS) are pivotal component in the energy market, serving as both energy suppliers and consumers. ESS operators can reap benefits from energy arbitrage by optimizing operations of storage equipment. To further enhance ESS flexibility within the energy market and improve renewable energy utilization, a heterogeneous photovoltaic-ESS (PV-ESS) is proposed, which leverages the unique characteristics of battery energy storage (BES) and hydrogen energy storage (HES). For scheduling tasks of the heterogeneous PV-ESS, cost description plays a crucial role in guiding operator's strategies to maximize benefits. We develop a comprehensive cost function that takes into account degradation, capital, and operation/maintenance costs to reflect real-world scenarios. Moreover, while numerous methods excel in optimizing ESS energy arbitrage, they often rely on black-box models with opaque decision-making processes, limiting practical applicability. To overcome this limitation and enable transparent scheduling strategies, a prototype-based policy network with inherent interpretability is introduced. This network employs human-designed prototypes to guide decision-making by comparing similarities between prototypical situations and encountered situations, which allows for naturally explained scheduling strategies. Comparative results across four distinct cases underscore the effectiveness and practicality of our proposed pre-hoc interpretable optimization method when contrasted with black-box models.
We propose a framework for the design of feedback controllers that combines the optimization-driven and model-free advantages of deep reinforcement learning with the stability guarantees provided by using the Youla-Kucera parameterization to define the search domain. Recent advances in behavioral systems allow us to construct a data-driven internal model; this enables an alternative realization of the Youla-Kucera parameterization based entirely on input-output exploration data. Perhaps of independent interest, we formulate and analyze the stability of such data-driven models in the presence of noise. The Youla-Kucera approach requires a stable "parameter" for controller design. For the training of reinforcement learning agents, the set of all stable linear operators is given explicitly through a matrix factorization approach. Moreover, a nonlinear extension is given using a neural network to express a parameterized set of stable operators, which enables seamless integration with standard deep learning libraries. Finally, we show how these ideas can also be applied to tune fixed-structure controllers.
Deep reinforcement learning has achieved remarkable performance in various domains by leveraging deep neural networks for approximating value functions and policies. However, using neural networks to approximate value functions or policy functions still faces challenges, including low sample efficiency and overfitting. In this paper, we introduce OMNet, a novel learning paradigm utilizing multiple subnetworks within a single network, offering diverse outputs efficiently. We provide a systematic pipeline, including initialization, training, and sampling with OMNet. OMNet can be easily applied to various deep reinforcement learning algorithms with minimal additional overhead. Through comprehensive evaluations conducted on MuJoCo benchmark, our findings highlight OMNet's ability to strike an effective balance between performance and computational cost.
The pre-train and fine-tune paradigm in machine learning has had dramatic success in a wide range of domains because the use of existing data or pre-trained models on the internet enables quick and easy learning of new tasks. We aim to enable this paradigm in robotic reinforcement learning, allowing a robot to learn a new task with little human effort by leveraging data and models from the Internet. However, reinforcement learning often requires significant human effort in the form of manual reward specification or environment resets, even if the policy is pre-trained. We introduce RoboFuME, a reset-free fine-tuning system that pre-trains a multi-task manipulation policy from diverse datasets of prior experiences and self-improves online to learn a target task with minimal human intervention. Our insights are to utilize calibrated offline reinforcement learning techniques to ensure efficient online fine-tuning of a pre-trained policy in the presence of distribution shifts and leverage pre-trained vision language models (VLMs) to build a robust reward classifier for autonomously providing reward signals during the online fine-tuning process. In a diverse set of five real robot manipulation tasks, we show that our method can incorporate data from an existing robot dataset collected at a different institution and improve on a target task within as little as 3 hours of autonomous real-world experience. We also demonstrate in simulation experiments that our method outperforms prior works that use different RL algorithms or different approaches for predicting rewards. Project website: https://robofume.github.io
The primacy bias in model-free reinforcement learning (MFRL), which refers to the agent's tendency to overfit early data and lose the ability to learn from new data, can significantly decrease the performance of MFRL algorithms. Previous studies have shown that employing simple techniques, such as resetting the agent's parameters, can substantially alleviate the primacy bias in MFRL. However, the primacy bias in model-based reinforcement learning (MBRL) remains unexplored. In this work, we focus on investigating the primacy bias in MBRL. We begin by observing that resetting the agent's parameters harms its performance in the context of MBRL. We further find that the primacy bias in MBRL is more closely related to the primacy bias of the world model instead of the primacy bias of the agent. Based on this finding, we propose \textit{world model resetting}, a simple yet effective technique to alleviate the primacy bias in MBRL. We apply our method to two different MBRL algorithms, MBPO and DreamerV2. We validate the effectiveness of our method on multiple continuous control tasks on MuJoCo and DeepMind Control Suite, as well as discrete control tasks on Atari 100k benchmark. The experimental results show that \textit{world model resetting} can significantly alleviate the primacy bias in the model-based setting and improve the algorithm's performance. We also give a guide on how to perform \textit{world model resetting} effectively.
This paper presents the new Deep Reinforcement Learning (DRL) library RL-X and its application to the RoboCup Soccer Simulation 3D League and classic DRL benchmarks. RL-X provides a flexible and easy-to-extend codebase with self-contained single directory algorithms. Through the fast JAX-based implementations, RL-X can reach up to 4.5x speedups compared to well-known frameworks like Stable-Baselines3.
Identifying causal structure is central to many fields ranging from strategic decision-making to biology and economics. In this work, we propose CD-UCT, a model-based reinforcement learning method for causal discovery based on tree search that builds directed acyclic graphs incrementally. We also formalize and prove the correctness of an efficient algorithm for excluding edges that would introduce cycles, which enables deeper discrete search and sampling in DAG space. The proposed method can be applied broadly to causal Bayesian networks with both discrete and continuous random variables. We conduct a comprehensive evaluation on synthetic and real-world datasets, showing that CD-UCT substantially outperforms the state-of-the-art model-free reinforcement learning technique and greedy search, constituting a promising advancement for combinatorial methods.
As machine learning models become more capable, they have exhibited increased potential in solving complex tasks. One of the most promising directions uses deep reinforcement learning to train autonomous agents in computer network defense tasks. This work studies the impact of the reward signal that is provided to the agents when training for this task. Due to the nature of cybersecurity tasks, the reward signal is typically 1) in the form of penalties (e.g., when a compromise occurs), and 2) distributed sparsely across each defense episode. Such reward characteristics are atypical of classic reinforcement learning tasks where the agent is regularly rewarded for progress (cf. to getting occasionally penalized for failures). We investigate reward shaping techniques that could bridge this gap so as to enable agents to train more sample-efficiently and potentially converge to a better performance. We first show that deep reinforcement learning algorithms are sensitive to the magnitude of the penalties and their relative size. Then, we combine penalties with positive external rewards and study their effect compared to penalty-only training. Finally, we evaluate intrinsic curiosity as an internal positive reward mechanism and discuss why it might not be as advantageous for high-level network monitoring tasks.
In multi-agent systems, agents need to interact and collaborate with other agents in environments. Agent modeling is crucial to facilitate agent interactions and make adaptive cooperation strategies. However, it is challenging for agents to model the beliefs, behaviors, and intentions of other agents in non-stationary environment where all agent policies are learned simultaneously. In addition, the existing methods realize agent modeling through behavior cloning which assume that the local information of other agents can be accessed during execution or training. However, this assumption is infeasible in unknown scenarios characterized by unknown agents, such as competition teams, unreliable communication and federated learning due to privacy concerns. To eliminate this assumption and achieve agent modeling in unknown scenarios, Fact-based Agent modeling (FAM) method is proposed in which fact-based belief inference (FBI) network models other agents in partially observable environment only based on its local information. The reward and observation obtained by agents after taking actions are called facts, and FAM uses facts as reconstruction target to learn the policy representation of other agents through a variational autoencoder. We evaluate FAM on various Multiagent Particle Environment (MPE) and compare the results with several state-of-the-art MARL algorithms. Experimental results show that compared with baseline methods, FAM can effectively improve the efficiency of agent policy learning by making adaptive cooperation strategies in multi-agent reinforcement learning tasks, while achieving higher returns in complex competitive-cooperative mixed scenarios.
Offline reinforcement learning (RL) presents a promising approach for learning reinforced policies from offline datasets without the need for costly or unsafe interactions with the environment. However, datasets collected by humans in real-world environments are often noisy and may even be maliciously corrupted, which can significantly degrade the performance of offline RL. In this work, we first investigate the performance of current offline RL algorithms under comprehensive data corruption, including states, actions, rewards, and dynamics. Our extensive experiments reveal that implicit Q-learning (IQL) demonstrates remarkable resilience to data corruption among various offline RL algorithms. Furthermore, we conduct both empirical and theoretical analyses to understand IQL's robust performance, identifying its supervised policy learning scheme as the key factor. Despite its relative robustness, IQL still suffers from heavy-tail targets of Q functions under dynamics corruption. To tackle this challenge, we draw inspiration from robust statistics to employ the Huber loss to handle the heavy-tailedness and utilize quantile estimators to balance penalization for corrupted data and learning stability. By incorporating these simple yet effective modifications into IQL, we propose a more robust offline RL approach named Robust IQL (RIQL). Extensive experiments demonstrate that RIQL exhibits highly robust performance when subjected to diverse data corruption scenarios.