19,991 Hits in 6.3 sec

Reinforcement Learning for Multi-Objective Optimization of Online Decisions in High-Dimensional Systems [article]

Hardik Meisheri and Vinita Baniwal and Nazneen N Sultana and Balaraman Ravindran and Harshad Khadilkar
2019 arXiv   pre-print
This paper describes a purely data-driven solution to a class of sequential decision-making problems with a large number of concurrent online decisions, with applications to computing systems and operations  ...  We first formulate the decision-making problem as a canonical reinforcement learning (RL) problem, which can be solved using purely data-driven techniques.  ...  Acknowledgment The authors would like to thank Dheeraj Shah and Padmakumar Ma from TCS Retail Strategic Initiatives team, for their help in defining the inventory management use case.  ... 
arXiv:1910.00211v1 fatcat:akkf6mny4vecnnypmmc7oxwctq

Multi-Task Generative Adversarial Nets with Shared Memory for Cross-Domain Coordination Control [article]

JunPing Wang, WenSheng Zhang, Ian Thomas, ShiHui Duan, YouKang Shi
2018 arXiv   pre-print
all of tasks, and online evaluate performance of system actions in discrete-time nonlinear systems.  ...  Generating sequential decision process from huge amounts of measured process data is a future research direction for collaborative factory automation, making full use of those online or offline process  ...  This research has been supported by the Project for Natural Science Foundation of China No.U1636220, No.61772525.  ... 
arXiv:1807.00298v1 fatcat:we2c24rbnrfkbeanpjdqihxljq

Reinforcement Learning for Online Information Seeking [article]

Xiangyu Zhao and Long Xia and Jiliang Tang and Dawei Yin
2019 arXiv   pre-print
In this paper, we give an overview of deep reinforcement learning for search, recommendation, and online advertising from methodologies to applications, review representative algorithms, and discuss some  ...  With recent great advances in deep reinforcement learning (DRL), there have been increasing interests in developing DRL based information seeking techniques.  ...  Policy Learning Reinforcement Learning is a class of learning problems in which the goal of an agent (or multi-agent) to find the policy to optimize some measures of its long-term performance.  ... 
arXiv:1812.07127v4 fatcat:pyc75g5hufcs5b3f75gonbkp24

gTLO: A Generalized and Non-linear Multi-Objective Deep Reinforcement Learning Approach [article]

Johannes Dornheim
2022 arXiv   pre-print
In the case of multi-policy MORL, sets of decision policies for various preferences regarding the conflicting objectives are optimized.  ...  In contrast, multi-objective reinforcement learning (MORL) methods learn from vectors of per-objective rewards instead.  ...  Multi-Objective Reinforcement Learning In many decision optimization problems, instead of a single objective, multiple objectives have to be optimized simultaneously.  ... 
arXiv:2204.04988v1 fatcat:dj7fniwe6rfx5h2b4ulya5i7ii

Toward Experience-Driven Traffic Management and Orchestration in Digital-Twin-Enabled 6G Networks [article]

Muhammad Tariq, Faisal Naeem, H. Vincent Poor
2022 arXiv   pre-print
, terahertz and millimeter communication), computing systems (e.g., cloud computing and fog computing) with its associated algorithms (e.g., optimization and machine learning).  ...  The envisioned 6G networks are expected to support extremely high data rates, low-latency, and radically new applications empowered by machine learning.  ...  performance of the network. 6) Multi-agent reinforcement learning for envisioned 6G networks: The multi-agent reinforcement learning (MARL) is a recent technique that involves set of agents and the objective  ... 
arXiv:2201.04259v1 fatcat:vi4bat522zfftpm5b4snnarbq4

Optimizing Online Matching for Ride-Sourcing Services with Multi-Agent Deep Reinforcement Learning [article]

Jintao Ke, Feng Xiao, Hai Yang, Jieping Ye
2019 arXiv   pre-print
Online matching between idle drivers and waiting passengers is one of the most key components in a ride-sourcing system.  ...  Motivated by the potential benefits of delayed matching, this paper establishes a two-stage framework which incorporates a combinatorial optimization and multi-agent deep reinforcement learning methods  ...  ACKNOWLEDGMENTS The work described in this paper was supported by Hong Kong Research Grants Council under projects HKUST16222916, NHKUST627/18 and the National Natural Science Foundation of China under  ... 
arXiv:1902.06228v1 fatcat:uwl2xwgeffcd5jwojjggm4yzim

Review, Analyze, and Design a Comprehensive Deep Reinforcement Learning Framework [article]

Ngoc Duy Nguyen, Thanh Thi Nguyen, Hai Nguyen, Saeid Nahavandi
2020 arXiv   pre-print
More importantly, there has been a great attention to RL since the introduction of deep learning that essentially makes RL feasible to operate in high-dimensional environments.  ...  Reinforcement learning (RL) has emerged as a standard approach for building an intelligent system, which involves multiple self-operated agents to collectively accomplish a designated task.  ...  Multi-objective environment and Use a multi-objective learner (MO Q-Learning) fruit/samples/basic/multi objectives multi-objective RL to train an agent to play Mountain Car [47] 5.  ... 
arXiv:2002.11883v1 fatcat:yziq6kwryvh5hiwjm6ju2r5srq

An Introduction to Deep Reinforcement Learning

Vincent François-Lavet, Peter Henderson, Riashat Islam, Marc G. Bellemare, Joelle Pineau
2018 Foundations and Trends® in Machine Learning  
Deep reinforcement learning is the combination of reinforcement learning (RL) and deep learning.  ...  This field of research has been able to solve a wide range of complex decision-making tasks that were previously out of reach for a machine.  ...  ISSN online version 1935 . Also available as a combined paper and online subscription. Introduction Motivation A core topic in machine learning is that of sequential decision-making.  ... 
doi:10.1561/2200000071 fatcat:gh3odyludnc43oeiqrgrtaer3u

Application of Machine Learning and Rule Scheduling in a Job-Shop Production Control System

Y. Zhao, H. Zhang
2021 International Journal of Simulation Modelling  
Firstly, a multi-objective optimization model was established for the production control system of dynamic job-shop.  ...  Then, deep reinforcement learning was introduced to job-shop production control system to transform the dynamic job-shop production control problem.  ...  Figure 1 : 1 Multi-objective optimization model for the production control system in a dynamic job-shop.  ... 
doi:10.2507/ijsimm20-2-co10 fatcat:anlk7inkpvbexi67jd7m6wqv4e

Multi-Agent Deep Reinforcement Learning for Multi-Object Tracker

Mingxin Jiang, Chao Deng, Yinshan Yu, Jingsong Shan
2019 IEEE Access  
We propose a novel approach based on multi-agent deep reinforcement learning (MADRL) for multi-object tracking to solve the problems in the existing tracking methods, such as a varying number of targets  ...  Then, we conducted offline learning in the training and online learning during the tracking.  ...  Compact low-dimensional features of high-dimensional data (such as images, text, and audio) can be found by deep neural networks automatically, which is the most outstanding contribution of deep learning  ... 
doi:10.1109/access.2019.2901300 fatcat:gb5n4dhnaba5rbtyyf45gz5q7i

Model-free Deep Reinforcement Learning for Urban Autonomous Driving [article]

Jianyu Chen, Bodi Yuan, Masayoshi Tomizuka
2019 arXiv   pre-print
In this paper, we propose a framework to enable model-free deep reinforcement learning in challenging urban autonomous driving scenarios.  ...  Urban autonomous driving decision making is challenging due to complex road geometry and multi-agent interactions.  ...  INTRODUCTION A highly intelligent decision making system is crucial for urban autonomous driving with dense surrounding dynamic objects.  ... 
arXiv:1904.09503v2 fatcat:wxfrqwuowvhihh3h75mixlcyie

Reinforcement Learning in Dynamic Task Scheduling: A Review

Chathurangi Shyalika, Thushari Silva, Asoka Karunananda
2020 SN Computer Science  
Reinforcement Learning is an emergent technology which has been able to solve the problem of the optimal task and resource scheduling dynamically.  ...  The paper addresses the results of the study by means of the state-of-theart on Reinforcement learning techniques used in dynamic task scheduling and a comparative review of those techniques.  ...  Thushari Silva and Professor Asoka Karunananda for their massive guidance and commitment throughout the research. Funding The funding is handled by the Authors itself.  ... 
doi:10.1007/s42979-020-00326-5 fatcat:egp6vgpetbcwdasm45vunmo3n4

Offline Reinforcement Learning for Mobile Notifications [article]

Yiping Yuan, Ajith Muralidharan, Preetam Nandy, Miao Cheng, Prakruthi Prabhakar
2022 arXiv   pre-print
We propose an offline reinforcement learning framework to optimize sequential notification decisions for driving user engagement.  ...  In this paper, we argue that reinforcement learning is a better framework for notification systems in terms of performance and iteration speed.  ...  feedback during the development of this work.  ... 
arXiv:2202.03867v1 fatcat:v4yibo6htvc6jbwgh7s4f5rhyq

Guest Editorial: Introduction to the Special Section on Machine Learning-Based Internet of Vehicles: Theory, Methodology, and Applications

Jun Guo, Sunwoo Kim, Henk Wymeersch, Walid Saad, Wei Chen
2019 IEEE Transactions on Vehicular Technology  
They formulate the offloading decision of multi-task in a service as a long-term planning problem, and explores the recent deep reinforcement learning to obtain the optimal solution.  ...  Decision for Vehicular Edge Computing: A Deep Reinforcement Learning Approach," which provides the optimal policy directly from the environment.  ... 
doi:10.1109/tvt.2019.2914747 fatcat:rrpckr7cczfdzmqy7nkbcnsdua

A Function Approximation Method for Model-based High-Dimensional Inverse Reinforcement Learning [article]

Kun Li, Joel W. Burdick
2017 arXiv   pre-print
This works handles the inverse reinforcement learning problem in high-dimensional state spaces, which relies on an efficient solution of model-based high-dimensional reinforcement learning problems.  ...  based on the observed human actions for inverse reinforcement learning problems.  ...  The method in [22] only learns an optimal value function, instead of the reward function. III. HIGH-DIMENSIONAL INVERSE REINFORCEMENT LEARNING A.  ... 
arXiv:1708.07738v1 fatcat:nvqbhwqklfes7jd5zc5l5unvr4
« Previous Showing results 1 — 15 out of 19,991 results