arXiv subjects

Nathan P. Lawrence

Publications and source records attributed to Nathan P. Lawrence.

At least 19 recordsLinked to original sources

arXiv2026

Error whitening: Why Gauss-Newton outperforms Newton

The Gauss-Newton matrix is widely viewed as a positive semidefinite approximation of the Hessian, yet mounting empirical evidence shows that Gauss-Newton descent outperforms Newton's method. We adopt a function space perspective to analyze this phenomenon. We show that the generalized Gauss-Newton (GGN) matrix projects the Newton direction in function space onto the model's tangent space, while a Jacobian-only variant obtained by applying the least squares Gauss-Newton matrix to non-least squares losses projects the function space loss gradient onto this same tangent space. Both projections eliminate distortions from the model's parameterization. Specifically, the evolution of the prediction-target mismatch depends on the model's parameterization through the matrix $JJ^\top$ where $J$ is the Jacobian of the model with respect to its parameters. The projections effectively replace $JJ^\top$ with the identity. We call this effect error whitening. Once the parameterization is removed, the prediction-target mismatch evolves according to dynamics dictated by the structure of the loss and the projection produced by the optimizer. Error whitening is a special property of Gauss-Newton descent that rigorously distinguishes it from Newton's method. We empirically demonstrate that Gauss-Newton optimizers follow the theoretically predicted function space dynamics and outperforms Newton's method, Adam, and Muon across case studies spanning supervised learning, physics-informed deep learning, and approximate dynamic programming.

cs.LG↗

arXiv2026

The Separation Principle and the Dual-Certainty Equivalence Gap in Model Predictive Control

Dual control addresses the trade-off between exploitation and exploration, where control inputs both regulate the system and generate informative data for estimation and identification. For certain problem classes, control and estimation can be designed independently without loss of optimality, a property known as the separation principle. However, in stochastic control problems with model uncertainty and constraints, this principle generally breaks down, and introduces the need for dual control. In this paper, we propose an information-weighted dual model predictive control (MPC) formulation and introduce metrics that quantify the dependence of the MPC policy on the uncertainty. We focus on parametric uncertainty in linear systems with Gaussian noise, though the metrics can be applied more broadly. Numerical results show that the dependence of the MPC policy on the posterior covariance is largest under high uncertainty and vanishes as the posterior covariance contracts, providing empirical evidence of the dual effect in closed loop. Moreover, the dual controller improves regulation performance and model accuracy compared to certainty-equivalent MPC.

math.OC↗

arXiv2026

Soft MPCritic: Amortized Model Predictive Value Iteration

Reinforcement learning (RL) and model predictive control (MPC) offer complementary strengths, yet combining them at scale remains computationally challenging. We propose soft MPCritic, an RL-MPC framework that learns in (soft) value space while using sample-based planning for both online control and value target generation. soft MPCritic instantiates MPC through model predictive path integral control (MPPI) and trains a terminal Q-function with fitted value iteration, aligning the learned value function with the planner and implicitly extending the effective planning horizon. We introduce an amortized warm-start strategy that recycles planned open-loop action sequences from online observations when computing batched MPPI-based value targets. This makes soft MPCritic computationally practical, while preserving solution quality. soft MPCritic plans in a scenario-based fashion with an ensemble of dynamic models trained for next-step prediction accuracy. Together, these ingredients enable soft MPCritic to learn effectively through robust, short-horizon planning on classic and complex control tasks. These results establish soft MPCritic as a practical and scalable blueprint for synthesizing MPC policies in settings where policy extraction and direct, long-horizon planning may fail.

cs.LG↗

arXiv2025

Why Goal-Conditioned Reinforcement Learning Works: Relation to Dual Control

Goal-conditioned reinforcement learning (RL) concerns the problem of training an agent to maximize the probability of reaching target goal states. This paper presents an analysis of the goal-conditioned setting based on optimal control. In particular, we derive an optimality gap between more classical, often quadratic, objectives and the goal-conditioned reward, elucidating the success of goal-conditioned RL and why classical ``dense'' rewards can falter. We then consider the partially observed Markov decision setting and connect state estimation to our probabilistic reward, making the goal-conditioned reward well suited to dual control problems. The advantages of goal-conditioned policies are validated on nonlinear and uncertain environments using both RL and predictive control techniques.

cs.LG↗

arXiv2025

A view on learning robust goal-conditioned value functions: Interplay between RL and MPC

Reinforcement learning (RL) and model predictive control (MPC) offer a wealth of distinct approaches for automatic decision-making under uncertainty. Given the impact both fields have had independently across numerous domains, there is growing interest in combining the general-purpose learning capability of RL with the safety and robustness features of MPC. To this end, this paper presents a tutorial-style treatment of RL and MPC, treating them as alternative approaches to solving Markov decision processes. In our formulation, RL aims to learn a global value function through offline exploration in an uncertain environment, whereas MPC constructs a local value function through online optimization. This local-global perspective suggests new ways to design policies that combine robustness and goal-conditioned learning. Robustness is incorporated into the RL and MPC pipelines through a scenario-based approach. Goal-conditioned learning aims to alleviate the burden of engineering a reward function for RL. Combining the two leads to a single policy that unites a robust, high-level RL terminal value function with short-term, scenario-based MPC planning for reliable constraint satisfaction. This approach leverages the benefits of both RL and MPC, the effectiveness of which is demonstrated on classical control benchmarks.

eess.SY↗

arXiv2025

MPCritic: A plug-and-play MPC architecture for reinforcement learning

The reinforcement learning (RL) and model predictive control (MPC) communities have developed vast ecosystems of theoretical approaches and computational tools for solving optimal control problems. Given their conceptual similarities but differing strengths, there has been increasing interest in synergizing RL and MPC. However, existing approaches tend to be limited for various reasons, including computational cost of MPC in an RL algorithm and software hurdles towards seamless integration of MPC and RL tools. These challenges often result in the use of "simple" MPC schemes or RL algorithms, neglecting the state-of-the-art in both areas. This paper presents MPCritic, a machine learning-friendly architecture that interfaces seamlessly with MPC tools. MPCritic utilizes the loss landscape defined by a parameterized MPC problem, focusing on "soft" optimization over batched training steps; thereby updating the MPC parameters while avoiding costly minimization and parametric sensitivities. Since the MPC structure is preserved during training, an MPC agent can be readily used for online deployment, where robust constraint satisfaction is paramount. We demonstrate the versatility of MPCritic, in terms of MPC architectures and RL algorithms that it can accommodate, on classic control benchmarks.

cs.LG↗

arXiv2025

ARMOR: Robust Reinforcement Learning-based Control for UAVs under Physical Attacks

Unmanned Aerial Vehicles (UAVs) depend on onboard sensors for perception, navigation, and control. However, these sensors are susceptible to physical attacks, such as GPS spoofing, that can corrupt state estimates and lead to unsafe behavior. While reinforcement learning (RL) offers adaptive control capabilities, existing safe RL methods are ineffective against such attacks. We present ARMOR (Adaptive Robust Manipulation-Optimized State Representations), an attack-resilient, model-free RL controller that enables robust UAV operation under adversarial sensor manipulation. Instead of relying on raw sensor observations, ARMOR learns a robust latent representation of the UAV's physical state via a two-stage training framework. In the first stage, a teacher encoder, trained with privileged attack information, generates attack-aware latent states for RL policy training. In the second stage, a student encoder is trained via supervised learning to approximate the teacher's latent states using only historical sensor data, enabling real-world deployment without privileged information. Our experiments show that ARMOR outperforms conventional methods, ensuring UAV safety. Additionally, ARMOR improves generalization to unseen attacks and reduces training cost by eliminating the need for iterative adversarial training.

cs.LG↗

arXiv2025

Local-Global Learning of Interpretable Control Policies: The Interface between MPC and Reinforcement Learning

Making optimal decisions under uncertainty is a shared problem among distinct fields. While optimal control is commonly studied in the framework of dynamic programming, it is approached with differing perspectives of the Bellman optimality condition. In one perspective, the Bellman equation is used to derive a global optimality condition useful for iterative learning of control policies through interactions with an environment. Alternatively, the Bellman equation is also widely adopted to derive tractable optimization-based control policies that satisfy a local notion of optimality. By leveraging ideas from the two perspectives, we present a local-global paradigm for optimal control suited for learning interpretable local decision makers that approximately satisfy the global Bellman equation. The benefits and practical complications in local-global learning are discussed. These aspects are exemplified through case studies, which give an overview of two distinct strategies for unifying reinforcement learning and model predictive control. We discuss the challenges and trade-offs in these local-global strategies, towards highlighting future research opportunities for safe and optimal decision-making under uncertainty.

eess.SY↗

arXiv2024

Guiding Reinforcement Learning with Incomplete System Dynamics

Model-free reinforcement learning (RL) is inherently a reactive method, operating under the assumption that it starts with no prior knowledge of the system and entirely depends on trial-and-error for learning. This approach faces several challenges, such as poor sample efficiency, generalization, and the need for well-designed reward functions to guide learning effectively. On the other hand, controllers based on complete system dynamics do not require data. This paper addresses the intermediate situation where there is not enough model information for complete controller design, but there is enough to suggest that a model-free approach is not the best approach either. By carefully decoupling known and unknown information about the system dynamics, we obtain an embedded controller guided by our partial model and thus improve the learning efficiency of an RL-enhanced approach. A modular design allows us to deploy mainstream RL algorithms to refine the policy. Simulation results show that our method significantly improves sample efficiency compared with standard RL methods on continuous control tasks, and also offers enhanced performance over traditional control approaches. Experiments on a real ground vehicle also validate the performance of our method, including generalization and robustness.

cs.RO↗

arXiv2024

Deep Hankel matrices with random elements

Willems' fundamental lemma enables a trajectory-based characterization of linear systems through data-based Hankel matrices. However, in the presence of measurement noise, we ask: Is this noisy Hankel-based model expressive enough to re-identify itself? In other words, we study the output prediction accuracy from recursively applying the same persistently exciting input sequence to the model. We find an asymptotic connection to this self-consistency question in terms of the amount of data. More importantly, we also connect this question to the depth (number of rows) of the Hankel model, showing the simple act of reconfiguring a finite dataset significantly improves accuracy. We apply these insights to find a parsimonious depth for LQR problems over the trajectory space.

eess.SY↗

arXiv2024

Stabilizing reinforcement learning control: A modular framework for optimizing over all stable behavior

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.

cs.LG↗

arXiv2024

Machine learning for industrial sensing and control: A survey and practical perspective

With the rise of deep learning, there has been renewed interest within the process industries to utilize data on large-scale nonlinear sensing and control problems. We identify key statistical and machine learning techniques that have seen practical success in the process industries. To do so, we start with hybrid modeling to provide a methodological framework underlying core application areas: soft sensing, process optimization, and control. Soft sensing contains a wealth of industrial applications of statistical and machine learning methods. We quantitatively identify research trends, allowing insight into the most successful techniques in practice. We consider two distinct flavors for data-driven optimization and control: hybrid modeling in conjunction with mathematical programming techniques and reinforcement learning. Throughout these application areas, we discuss their respective industrial requirements and challenges. A common challenge is the interpretability and efficiency of purely data-driven methods. This suggests a need to carefully balance deep learning techniques with domain knowledge. As a result, we highlight ways prior knowledge may be integrated into industrial machine learning applications. The treatment of methods, problems, and applications presented here is poised to inform and inspire practitioners and researchers to develop impactful data-driven sensing, optimization, and control solutions in the process industries.

eess.SY↗

arXiv2023

Reinforcement Learning with Partial Parametric Model Knowledge

We adapt reinforcement learning (RL) methods for continuous control to bridge the gap between complete ignorance and perfect knowledge of the environment. Our method, Partial Knowledge Least Squares Policy Iteration (PLSPI), takes inspiration from both model-free RL and model-based control. It uses incomplete information from a partial model and retains RL's data-driven adaption towards optimal performance. The linear quadratic regulator provides a case study; numerical experiments demonstrate the effectiveness and resulting benefits of the proposed method.

eess.SY↗

arXiv2023

A modular framework for stabilizing deep reinforcement learning control

eess.SY↗

arXiv2023

Automated deep reinforcement learning for real-time scheduling strategy of multi-energy system integrated with post-carbon and direct-air carbon captured system

The carbon-capturing process with the aid of CO2 removal technology (CDRT) has been recognised as an alternative and a prominent approach to deep decarbonisation. However, the main hindrance is the enormous energy demand and the economic implication of CDRT if not effectively managed. Hence, a novel deep reinforcement learning agent (DRL), integrated with an automated hyperparameter selection feature, is proposed in this study for the real-time scheduling of a multi-energy system coupled with CDRT. Post-carbon capture systems (PCCS) and direct-air capture systems (DACS) are considered CDRT. Various possible configurations are evaluated using real-time multi-energy data of a district in Arizona and CDRT parameters from manufacturers' catalogues and pilot project documentation. The simulation results validate that an optimised soft-actor critic (SAC) algorithm outperformed the TD3 algorithm due to its maximum entropy feature. We then trained four (4) SAC agents, equivalent to the number of considered case studies, using optimised hyperparameter values and deployed them in real time for evaluation. The results show that the proposed DRL agent can meet the prosumers' multi-energy demand and schedule the CDRT energy demand economically without specified constraints violation. Also, the proposed DRL agent outperformed rule-based scheduling by 23.65%. However, the configuration with PCCS and solid-sorbent DACS is considered the most suitable configuration with a high CO2 captured-released ratio of 38.54, low CO2 released indicator value of 2.53, and a 36.5% reduction in CDR cost due to waste heat utilisation and high absorption capacity of the selected sorbent. However, the adoption of CDRT is not economically viable at the current carbon price. Finally, we showed that CDRT would be attractive at a carbon price of 400-450USD/ton with the provision of tax incentives by the policymakers.

eess.SY↗

arXiv2022

Modern Machine Learning Tools for Monitoring and Control of Industrial Processes: A Survey

Over the last ten years, we have seen a significant increase in industrial data, tremendous improvement in computational power, and major theoretical advances in machine learning. This opens up an opportunity to use modern machine learning tools on large-scale nonlinear monitoring and control problems. This article provides a survey of recent results with applications in the process industry.

cs.LG↗

arXiv2022

Meta-Reinforcement Learning for Adaptive Control of Second Order Systems

Meta-learning is a branch of machine learning which aims to synthesize data from a distribution of related tasks to efficiently solve new ones. In process control, many systems have similar and well-understood dynamics, which suggests it is feasible to create a generalizable controller through meta-learning. In this work, we formulate a meta reinforcement learning (meta-RL) control strategy that takes advantage of known, offline information for training, such as a model structure. The meta-RL agent is trained over a distribution of model parameters, rather than a single model, enabling the agent to automatically adapt to changes in the process dynamics while maintaining performance. A key design element is the ability to leverage model-based information offline during training, while maintaining a model-free policy structure for interacting with new environments. Our previous work has demonstrated how this approach can be applied to the industrially-relevant problem of tuning proportional-integral controllers to control first order processes. In this work, we briefly reintroduce our methodology and demonstrate how it can be extended to proportional-integral-derivative controllers and second order systems.

cs.LG↗

arXiv2022

Meta-Reinforcement Learning for the Tuning of PI Controllers: An Offline Approach

Meta-learning is a branch of machine learning which trains neural network models to synthesize a wide variety of data in order to rapidly solve new problems. In process control, many systems have similar and well-understood dynamics, which suggests it is feasible to create a generalizable controller through meta-learning. In this work, we formulate a meta reinforcement learning (meta-RL) control strategy that can be used to tune proportional--integral controllers. Our meta-RL agent has a recurrent structure that accumulates "context" to learn a system's dynamics through a hidden state variable in closed-loop. This architecture enables the agent to automatically adapt to changes in the process dynamics. In tests reported here, the meta-RL agent was trained entirely offline on first order plus time delay systems, and produced excellent results on novel systems drawn from the same distribution of process dynamics used for training. A key design element is the ability to leverage model-based information offline during training in simulated environments while maintaining a model-free policy structure for interacting with novel processes where there is uncertainty regarding the true process dynamics. Meta-learning is a promising approach for constructing sample-efficient intelligent controllers.

eess.SY↗