paper-with-me

홈 › Papers

SwiftRL: Towards Efficient Reinforcement Learning on Real Processing-In-Memory Systems

2024-05-07 · Kailash Gogineni, Sai Santosh Dayapule, Juan Gómez-Luna, Karthikeya Gogineni, Peng Wei, Tian Lan, Mohammad Sadrosadati, Onur Mutlu, Guru Venkataramani

Reinforcement Learning (RL) trains agents to learn optimal behavior by maximizing reward signals from experience datasets. However, RL training often faces memory limitations, leading to execution latencies and prolonged training times. To overcome this, SwiftRL explores Processing-In-Memory (PIM) architectures to accelerate RL workloads. We achieve near-linear performance scaling by implementing RL algorithms like Tabular Q-learning and SARSA on UPMEM PIM systems and optimizing for hardware. Our experiments on OpenAI GYM environments using UPMEM hardware demonstrate superior performance compared to CPU and GPU implementations.

📄 PDF Abstract BibTeX arXiv:2405.03967

Code (1)

kailashg26/swiftrl 공식 구현

Tasks

CPUGPUOpenAI GymQ-Learningreinforcement-learningReinforcement Learning (RL)

Methods 이 논문이 사용한 방법론

Q-Learning Q-Learning is an off-policy temporal difference control algorithm: $$Q\left(S\_{t}, A\_{t}\right) \leftarrow Q\left(S\_{t}, A\_{t}\right) + \alpha\left[R_{t+1} +…
Sarsa Sarsa is an on-policy TD control algorithm: $$Q\left(S\_{t}, A\_{t}\right) \leftarrow Q\left(S\_{t}, A\_{t}\right) + \alpha\left[R_{t+1} + \gamma{Q}\left(S\_{t+1},…

Similar Papers 제목 키워드 기반

PyGim: An Efficient Graph Neural Network Library for Real Processing-In-Memory Architectures

2024-02-26 · Christina Giannoula, Peiming Yang, Ivan Fernandez, Jiacheng Yang 외

Graph Neural Networks (GNNs) are emerging ML models to analyze graph-structure data. Graph Neural Network (GNN) execution involves both compute-intensive and memory-intensive kernels, the latter dominates the total time,…

CPUGPUGraph Neural Network

Machine Learning Training on a Real Processing-in-Memory System

2022-06-13 · Juan Gómez-Luna, Yuxin Guo, Sylvan Brocard, Julien Legriel 외

Training machine learning algorithms is a computationally intensive process, which is frequently memory-bound due to repeatedly accessing large training datasets. As a result, processor-centric systems (e.g., CPU, GPU) s…

BIG-bench Machine LearningCPUGPUregression

Compute-in-Memory Implementation of State Space Models for Event Sequence Processing

2025-11-17 · Xiaoyu Zhang, Mingtao Hu, Sen Lu, Soohyeon Kim 외 arxiv

State space models (SSMs) have recently emerged as a powerful framework for long sequence processing, outperforming traditional methods on diverse benchmarks. Fundamentally, SSMs can generalize both recurrent and convolu…

Event-based vision

Modelling Working Memory using Deep Recurrent Reinforcement Learning

2019-09-11 · NeurIPS Workshop Neuro_AI 2019 12 · Pravish Sainath, Pierre Bellec, Guillaume Lajoie

In cognitive systems, the role of a working memory is crucial for visual reasoning and decision making. Tremendous progress has been made in understanding the mechanisms of the human/animal working memory, as well as in …

Decision Makingreinforcement-learningReinforcement LearningReinforcement Learning (RL)+1

Deep Reinforcement Learning: An Overview

2017-01-25 · Yuxi Li

We give an overview of recent exciting achievements of deep reinforcement learning (RL). We discuss six core elements, six important mechanisms, and twelve applications. We start with background of machine learning, deep…

Deep Reinforcement LearningMachine TranslationManagementreinforcement-learning+5