This paper presents Gym-TORAX, a Python package enabling the implementation of Reinforcement Learning (RL) environments for simulating plasma dynamics and control in tokamaks. Users define succinctly a set of control actions and observations, and a control objective from which Gym-TORAX creates a Gymnasium environment that wraps TORAX for simulating the plasma dynamics. The objective is formulated through rewards depending on the simulated state of the plasma and control action to optimize specific characteristics of the plasma, such as performance and stability. The resulting environment instance is then compatible with a wide range of RL algorithms and libraries and will facilitate RL research in plasma control. In its current version, one environment is readily available, based on a ramp-up scenario of the International Thermonuclear Experimental Reactor (ITER).

Note that this safe RL problem is less general than the standard formulation of safe RL. The authors introduce a teacher-student hierarchy. To learn the teacher's policy the following constraints are followed: a1 The unsafe set is contained in the intervention set D D The teacher learns when to intervene and to switch between different interventions. A1.2 RL with probability one constraints We have introduced the safety state to the environment as follows: s First, we discuss our design for the PI controller and discuss the necessary parts for it. The proportional part delivers brute force control by having a large control magnitude for large errors, but it is not effective if the instantaneous error values become small.

Learning from demonstrations is a powerful paradigm for robot manipulation, but its effectiveness hinges on both the quantity and quality of the collected data. In this work, we present a case study of how instrumentation, i.e. integration of sensors, can improve the quality of demonstrations and automate data collection. We instrument a squeeze bottle with a pressure sensor to learn a liquid dispensing task, enabling automated data collection via a PI controller. Transformer-based policies trained on automated demonstrations outperform those trained on human data in 78% of cases. Our findings indicate that instrumentation not only facilitates scalable data collection but also leads to better-performing policies, highlighting its potential in the pursuit of generalist robotic agents.

The deployment of neural networks on resource-constrained microcontrollers has gained momentum, driving many advancements in Tiny Neural Networks. This paper introduces a tiny feed-forward neural network, TinyFC, integrated into the Field-Oriented Control (FOC) of Permanent Magnet Synchronous Motors (PMSMs). Proportional-Integral (PI) controllers are widely used in FOC for their simplicity, although their limitations in handling nonlinear dynamics hinder precision. To address this issue, a lightweight 1,400 parameters TinyFC was devised to enhance the FOC performance while fitting into the computational and memory constraints of Figure 1: Workflow diagram to deploy NN-augmented FOC a micro-controller. Advanced optimization techniques, including pruning, hyperparameter tuning, and quantization to 8-bit integers, such as automotive, industrial, naval and aeronautics, where compact were applied to reduce the model's footprint while preserving the size and precision control are essential [19]. PMSMs consist of network effectiveness. Simulation results show the proposed approach a stator housing the windings and a rotor containing permanent significantly reduced overshoot by up to 87.5%, with the magnets. The operational interaction between the stator's rotating pruned model achieving complete overshoot elimination, highlighting magnetic field and the rotor's fixed magnetic field enables synchronization the potential of tiny neural networks in real-time motor control at synchronous speed [10].

Controller tuning is crucial for closed-loop performance but often involves manual adjustments. Although Bayesian optimization (BO) has been established as a data-efficient method for automated tuning, applying it to large and high-dimensional search spaces remains challenging. We extend a recently proposed local variant of BO to include crash constraints, where the controller can only be successfully evaluated in an a-priori unknown feasible region. We demonstrate the efficiency of the proposed method through simulations and hardware experiments. Our findings showcase the potential of local BO to enhance controller performance and reduce the time and resources necessary for tuning.

Constrained optimization offers a powerful framework to prescribe desired behaviors in neural network models. Typically, constrained problems are solved via their min-max Lagrangian formulations, which exhibit unstable oscillatory dynamics when optimized using gradient descent-ascent. The adoption of constrained optimization techniques in the machine learning community is currently limited by the lack of reliable, general-purpose update schemes for the Lagrange multipliers. This paper proposes the $\nu$PI algorithm and contributes an optimization perspective on Lagrange multiplier updates based on PI controllers, extending the work of Stooke, Achiam and Abbeel (2020). We provide theoretical and empirical insights explaining the inability of momentum methods to address the shortcomings of gradient descent-ascent, and contrast this with the empirical success of our proposed $\nu$PI controller. Moreover, we prove that $\nu$PI generalizes popular momentum methods for single-objective minimization. Our experiments demonstrate that $\nu$PI reliably stabilizes the multiplier dynamics and its hyperparameters enjoy robust and predictable behavior.

This paper presents a learning-based control strategy for non-linear throttle valves with an asymmetric hysteresis, leading to a near-optimal controller without requiring any prior knowledge about the environment. We start with a carefully tuned Proportional Integrator (PI) controller and exploit the recent advances in Reinforcement Learning (RL) with Guides to improve the closed-loop behavior by learning from the additional interactions with the valve. We test the proposed control method in various scenarios on three different valves, all highlighting the benefits of combining both PI and RL frameworks to improve control performance in non-linear stochastic systems. In all the experimental test cases, the resulting agent has a better sample efficiency than traditional RL agents and outperforms the PI controller.

This paper introduces a novel approach to PDE boundary control design using neural operators to alleviate stop-and-go instabilities in congested traffic flow. Our framework leverages neural operators to design control strategies for traffic flow systems. The traffic dynamics are described by the Aw-Rascle-Zhang (ARZ) model, which comprises a set of second-order coupled hyperbolic partial differential equations (PDEs). Backstepping method is widely used for boundary control of such PDE systems. The PDE model-based control design can be time-consuming and require intensive depth of expertise since it involves constructing and solving backstepping control kernels. To overcome these challenges, we present two distinct neural operator (NO) learning schemes aimed at stabilizing the traffic PDE system. The first scheme embeds NO-approximated gain kernels within a predefined backstepping controller, while the second one directly learns a boundary control law. The Lyapunov analysis is conducted to evaluate the stability of the NO-approximated gain kernels and control law. It is proved that the NO-based closed-loop system is practical stable under certain approximation accuracy conditions in NO-learning. To validate the efficacy of the proposed approach, simulations are conducted to compare the performance of the two neural operator controllers with a PDE backstepping controller and a Proportional Integral (PI) controller. While the NO-approximated methods exhibit higher errors compared to the backstepping controller, they consistently outperform the PI controller, demonstrating faster computation speeds across all scenarios. This result suggests that neural operators can significantly expedite and simplify the process of obtaining boundary controllers in traffic PDE systems.

As Exascale computing becomes a reality, the energy needs of compute nodes in cloud data centers will continue to grow. A common approach to reducing this energy demand is to limit the power consumption of hardware components when workloads are experiencing bottlenecks elsewhere in the system. However, designing a resource controller capable of detecting and limiting power consumption on-the-fly is a complex issue and can also adversely impact application performance. In this paper, we explore the use of Reinforcement Learning (RL) to design a power capping policy on cloud compute nodes using observations on current power consumption and instantaneous application performance (heartbeats). By leveraging the Argo Node Resource Management (NRM) software stack in conjunction with the Intel Running Average Power Limit (RAPL) hardware control mechanism, we design an agent to control the maximum supplied power to processors without compromising on application performance. Employing a Proximal Policy Optimization (PPO) agent to learn an optimal policy on a mathematical model of the compute nodes, we demonstrate and evaluate using the STREAM benchmark how a trained agent running on actual hardware can take actions by balancing power consumption and application performance.