An official website of the United States government
Actor frameworks and similar reactive programming techniques are widely used for building concurrent systems. They promise to be efficient and scale well to a large number of cores or nodes in a distributed system. However, they also expose programmers to nondeterminism, which often makes implementations hard to understand, debug, and test. The recently proposed reactor model is a promising alternative that enables deterministic concurrency. In this article, we present an efficient, parallel implementation of reactors and demonstrate that the determinacy of reactors does not imply a loss in performance. To show this, we evaluateLingua Franca(LF), a reactor-oriented coordination language. LF equips mainstream programming languages with a deterministic concurrency model that automatically takes advantage of opportunities to exploit parallelism. Our implementation of the Savina benchmark suite demonstrates that, in terms of execution time, the runtime performance of LF programs even exceeds popular and highly optimized actor frameworks. We compare against Akka and CAF, which LF outperforms by 1.86× and 1.42×, respectively.
more »
« less
Efficient Parallel Reinforcement Learning Framework Using the Reactor Model
Parallel Reinforcement Learning (RL) frameworks are essential for mapping RL workloads to multiple computational resources, allowing for faster generation of samples, estimation of values, and policy improvement. These computational paradigms require a seamless integration of training, serving, and simulation workloads. Existing frameworks, such as Ray, are not managing this orchestration efficiently, especially in RL tasks that demand intensive input/output and synchronization between actors on a single node. In this study, we have proposed a solution implementing the reactor model, which enforces a set of actors to have a fixed communication pattern. This allows the scheduler to eliminate work needed for synchronization, such as acquiring and releasing locks for each actor or sending and processing coordination-related messages. Our framework, Lingua Franca (LF), a coordination language based on the reactor model, also supports true parallelism in Python and provides a unified interface that allows users to automatically generate dataflow graphs for RL tasks. In comparison to Ray on a single-node multi-core compute platform, LF achieves 1.21x and 11.62x higher simulation throughput in OpenAI Gym and Atari environments, reduces the average training time of synchronized parallel Q-learning by 31.2%, and accelerates multi-agent RL inference by 5.12x.
more »
« less
- Award ID(s):
- 2233769
- PAR ID:
- 10554370
- Publisher / Repository:
- ACM
- Date Published:
- ISBN:
- 9798400704161
- Page Range / eLocation ID:
- 41 to 51
- Format(s):
- Medium: X
- Location:
- Nantes France
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Aleksandra Faust, David Hsu (Ed.)Modern Reinforcement Learning (RL) algorithms are not sample efficient to train on multi-step tasks in complex domains, impeding their wider deployment in the real world. We address this problem by leveraging the insight that RL models trained to complete one set of tasks can be repurposed to complete related tasks when given just a handful of demonstrations. Based upon this insight, we propose See-SPOT-Run (SSR), a new computational approach to robot learning that enables a robot to complete a variety of real robot tasks in novel problem domains without task-specific training. SSR uses pretrained RL models to create vectors that represent model, task, and action relevance in demonstration and test scenes. SSR then compares these vectors via our Cycle Consistency Distance (CCD) metric to determine the next action to take. SSR completes 58% more task steps and 20% more trials than a baseline few-shot learning method that requires task-specific training. SSR also achieves a four order of magnitude improvement in compute efficiency and a 20% to three order of magnitude improvement in sample efficiency compared to the baseline and to training RL models from scratch. To our knowledge, we are the first to address multi-step tasks from demonstration on a real robot without task-specific training, where both the visual input and action space output are high dimensional. Code is available in the supplement.more » « less
-
Distributed deep learning framework tools should aim at high efficiency of training and inference of distributed exascale deep learning algorithms. There are three major challenges in this endeavor: scalability, adaptivity and efficiency. Any future framework will need to be adaptively utilized for a variety of heterogeneous hardware and network environments and will thus be required to be capable of scaling from single compute node up to large clusters. Further, it should be efficiently integrated into popular frameworks such as TensorFlow, PyTorch, etc. This paper proposes a dynamically hybrid (hierarchy) distribution structure for distributed deep learning, taking advantage of flexible synchronization on both centralized and decentralized architectures, implementing multi-level fine-grain parallelism on distributed platforms. It is scalable as the number of compute nodes increases, and can also adapt to various compute abilities, memory structures and communication costs.more » « less
-
Recent advancements in Digital Twin (DT) technology have opened new avenues for smart manufacturing. These systems increasingly depend on adaptive control mechanisms to optimize complex processes and reduce production wastage. This research presents an innovative approach that integrates Soft Actor-Critic (SAC) Reinforcement Learning (RL) algorithm with DT technology with Robot Operating System 2 (ROS2) to enable real-time adaptive control in robotic manufacturing. Our experimental setup consists of a ViperX 300 S robot arm, in which two distinct Tasks: (1) static target reaching and (2) dynamic target following were implemented for simulating adaptive control of manufacturing process. The innovative system architecture combines Unity game engine’s simulation environment with ROS2 for seamless and robust DT synchronization. We implemented a hierarchical reward structure to address common RL challenges, including local minima avoidance, convergence acceleration, and training stability, while leveraging transfer learning to efficiently adapt trained behavior models across tasks. Experimental results demonstrate rapid policy convergence and robust task execution, with performance metrics including cumulative reward, value loss, policy loss, and entropy validating the effectiveness of the approach. To the best of our knowledge, this is the first study to integrate Unity with ROS2-based DT for real-time synchronization and adaptive physical robot control using RL. Unlike prior works limited to offline or low-frequency simulations, our framework achieves stable 20 ms joint-level synchronization, enabling deployment of learned behaviors directly to physical robotic systems through virtual platform. This work advances the integration of RL with realistic DT framework for industrial and manufacturing robotics applications, providing a framework for enhanced adaptive real-time control in smart additive manufacturing (AM) processes.more » « less
-
Recent advances in on-policy reinforcement learning (RL) methods enabled learning agents in virtual environments to master complex tasks with high-dimensional and continuous observation and action spaces. However, leveraging this family of algorithms in multi-fingered robotic grasping remains a challenge due to large sim-to-real fidelity gaps and the high sample complexity of on-policy RL algorithms. This work aims to bridge these gaps by first reinforcement-learning a multi-fingered robotic grasping policy in simulation that operates in the pixel space of the input: a single depth image. Using a mapping from pixel space to Cartesian space according to the depth map, this method transfers to the real world with high fidelity and introduces a novel attention mechanism that substantially improves grasp success rate in cluttered environments. Finally, the direct-generative nature of this method allows learning of multi-fingered grasps that have flexible end-effector positions, orientations and rotations, as well as all degrees of freedom of the hand.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1145/3626183.3659967
