Filters








1,244 Hits in 4.7 sec

Provably Efficient Multi-Task Reinforcement Learning with Model Transfer [article]

Chicheng Zhang, Zhi Wang
2022 arXiv   pre-print
We study multi-task reinforcement learning (RL) in tabular episodic Markov decision processes (MDPs).  ...  We design and analyze an algorithm based on the idea of model transfer, and provide gap-dependent and gap-independent upper and lower bounds that characterize the intrinsic complexity of the problem.  ...  We provide a provably efficient model-based algorithm that takes advantage of knowledge transfer between different tasks.  ... 
arXiv:2107.08622v2 fatcat:4ycn5hiyazd5djzm2rs55gphpm

Guest editor's introduction: special issue on inductive transfer learning

Daniel L. Silver, Kristin P. Bennett
2008 Machine Learning  
of related Reinforcement Learning (RL) tasks.  ...  The paper "Flexible latent variable models for multi-task learning" (Zhang et al. 2008 ) by Jian Zhang, Zoubin Ghahramani and Yiming Yang presents a Hierarchial Bayesian probabilistic framework for multi-task  ... 
doi:10.1007/s10994-008-5087-1 fatcat:icyvctwzyfds5iarryj7tkjinq

Reinforcement learning with multi-fidelity simulators

Mark Cutler, Thomas J. Walsh, Jonathan P. How
2014 2014 IEEE International Conference on Robotics and Automation (ICRA)  
We present a framework for reinforcement learning (RL) in a scenario where multiple simulators are available with decreasing amounts of fidelity to the real-world learning scenario.  ...  The approach allows RL algorithms to find nearoptimal policies for the real world with fewer expensive realworld samples than previous transfer approaches or learning without simulators.  ...  This paper introduces, analyzes, and empirically demonstrates a new framework, Multi-Fidelity Reinforcement Learning (MFRL), depicted in Figure 1 , for performing reinforcement learning with a heterogeneous  ... 
doi:10.1109/icra.2014.6907423 dblp:conf/icra/CutlerWH14 fatcat:gococffz7nbornymdwgcc2fuq4

Offline Pre-trained Multi-Agent Decision Transformer: One Big Sequence Model Tackles All SMAC Tasks [article]

Linghui Meng, Muning Wen, Yaodong Yang, Chenyang Le, Xiyun Li, Weinan Zhang, Ying Wen, Haifeng Zhang, Jun Wang, Bo Xu
2021 arXiv   pre-print
Such a paradigm is also desirable for multi-agent reinforcement learning (MARL) tasks, given the increased interactions among agents and with the enviroment.  ...  Offline reinforcement learning leverages previously-collected offline datasets to learn optimal policies with no necessity to access the real environment.  ...  In contrast, we show a transformer-based method in the multi-agent field, attempting to transfer across many scenarios without extra constraints. Multi-Agent Reinforcement Learning.  ... 
arXiv:2112.02845v2 fatcat:bkypmzhkpzbwfjlnve2ohqz5y4

Table of Contents

2021 IEEE Transactions on Vehicular Technology  
Aguilar 1255 Predictive Battery Health Management With Transfer Learning and Online Model Correction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  ...  Iizuka 1648 Multi-Agent Reinforcement Learning Based Distributed Transmission in Collaborative Cloud-Edge Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  ... 
doi:10.1109/tvt.2021.3058904 fatcat:oovl32jeojgo3iqw7nhpb6d3vy

Multi-Robot Coordination and Planning in Uncertain and Adversarial Environments [article]

Lifeng Zhou, Pratap Tokekar
2021 arXiv   pre-print
These algorithms have been applied to tasks such as formation control, task assignment and scheduling, search and planning, and informative data collection.  ...  In order for multi-robot systems to become practical, we need coordination algorithms that can scale to large teams of robots dealing with dynamically changing, failure-prone, contested, and uncertain  ...  Compliance with Ethical Standards Conflict of Interest The authors declare that they have no conflict of interest.  ... 
arXiv:2105.00389v1 fatcat:bwnyxvpvzjfrbjxyiysnmq5e74

Incremental reinforcement learning for designing multi-agent systems

Olivier Buffet, Alain Dutech, François Charpillet
2001 Proceedings of the fifth international conference on Autonomous agents - AGENTS '01  
One solution to automatically build such large Multi-Agent Systems is to use decentralized learning : each agent learns by itself its own behavior.  ...  To that purpose, Reinforcement Learning methods are very attractive as they do not require a solution of the problem to be known before hand.  ...  Furthermore, it is more efficient to learn a more complex task after an initial stage of incremental learning than learning directly this more complex task from scratch.  ... 
doi:10.1145/375735.375826 dblp:conf/agents/BuffetDC01 fatcat:qsczgp7lejhmfitkmpelopgz7q

Solving System Problems with Machine Learning

Ion STOICA
2019 Studies in Informatics and Control  
Over the past decade, Machine Learning (ML) has achieved tremendous successes and has seen wide-scale adoption for human-facing tasks, such as visual recognition, speech recognition, language translation  ...  One challenge is that solving many of these problems require solutions that are provably correct, which is at odds with the ML techniques which are stochastic in nature.  ...  Supervised learning Reinforcement Learning With reinforcement learning (RL) [43] , a software agent continuously interacts with the environment by taking actions.  ... 
doi:10.24846/v28i2y201901 fatcat:fktydwhmzzgvrexswwe246k4pm

Learning-Based Run-Time Power and Energy Management of Multi/Many-Core Systems: Current and Future Trends

Amit Kumar Singh, Charles Leech, Basireddy Karunakar Reddy, Bashir M. Al-Hashimi, Geoff V. Merrett
2017 Journal of Low Power Electronics  
These approaches perform design-time and/or run-time power/energy management by employing some learning principles such as reinforcement learning.  ...  Multi/Many-core systems are prevalent in several application domains targeting different scales of computing such as embedded and cloud computing.  ...  The traditional Q-learning algorithm provides a model-free solution for the MDP with a provable convergence to the optimal solution.  ... 
doi:10.1166/jolpe.2017.1492 fatcat:wrqqr345ujhglkpqt5hllonnxa

Combining Pessimism with Optimism for Robust and Efficient Model-Based Deep Reinforcement Learning [article]

Sebastian Curi, Ilija Bogunovic, Andreas Krause
2021 arXiv   pre-print
RH-UCRL is a model-based reinforcement learning (MBRL) algorithm that effectively distinguishes between epistemic and aleatoric uncertainty and efficiently explores both the agent and adversary decision  ...  In real-world tasks, reinforcement learning (RL) agents frequently encounter situations that are not present during training time.  ...  Data-efficient reinforcement learning with probabilistic model predictive control. In Conference on Artificial Intelligence and Statistics (AIS-TATS), pp. 1701-1710, 2018. Kingma, D. P. and Ba, J.  ... 
arXiv:2103.10369v1 fatcat:nqr7bcaugnfprolnvy5cq6c2iu

Deep Reinforcement Learning [article]

Yuxi Li
2018 arXiv   pre-print
We start with background of artificial intelligence, machine learning, deep learning, and reinforcement learning (RL), with resources.  ...  We discuss deep reinforcement learning in an overview style. We draw a big picture, filled with details.  ...  Lanctot et al. (2017) observe that independent RL, in which each agent learns by interacting with the environment, oblivious to other agents, can overfit the learned policies to other agents' policies  ... 
arXiv:1810.06339v1 fatcat:kp7atz5pdbeqta352e6b3nmuhy

More Than Privacy: Applying Differential Privacy in Key Areas of Artificial Intelligence [article]

Tianqing Zhu and Dayong Ye and Wei Wang and Wanlei Zhou and Philip S. Yu
2020 arXiv   pre-print
With a focus on regular machine learning, distributed machine learning, deep learning, and multi-agent systems, the purpose of this article is to deliver a new view on many possibilities for improving  ...  It can also be used to improve security, stabilize learning, build fair models, and impose composition in selected areas of AI.  ...  Differential privacy in multi-agent reinforcement learning Multi-agent learning is generally based on the reinforcement learning [103] .  ... 
arXiv:2008.01916v1 fatcat:ujmxv7eq6jcppndfu5shbzkdom

More Than Privacy: Applying Differential Privacy in Key Areas of Artificial Intelligence

Tianqing Zhu, Dayong Ye, Wei Wang, Wanlei Zhou, Philip Yu
2020 IEEE Transactions on Knowledge and Data Engineering  
With a focus on regular machine learning, distributed machine learning, deep learning, and multi-agent systems, the purpose of this article is to deliver a new view on many possibilities for improving  ...  It can also be used to improve security, stabilize learning, build fair models, and impose composition in selected areas of AI.  ...  Differential privacy in multi-agent reinforcement learning Multi-agent learning is generally based on the reinforcement learning [103] .  ... 
doi:10.1109/tkde.2020.3014246 fatcat:33rl6jxy5rgexpnuel5rvlkg5a

Real-World Reinforcement Learning via Multifidelity Simulators

Mark Cutler, Thomas J. Walsh, Jonathan P. How
2015 IEEE Transactions on robotics  
We present a framework for efficient RL in a scenario where multiple simulators of a target task are available, each with varying levels of fidelity.  ...  Reinforcement learning (RL) can be a tool for designing policies and controllers for robotic systems.  ...  In transfer learning (TL), values or model parameters are typically used to bootstrap learning in the next task, e.g. [19] .  ... 
doi:10.1109/tro.2015.2419431 fatcat:bmisjruwufhdlcfywvg5nnebgy

Context-Aware Policy Reuse [article]

Siyuan Li, Fangda Gu, Guangxiang Zhu, Chongjie Zhang
2019 arXiv   pre-print
Transfer learning can greatly speed up reinforcement learning for a new task by leveraging policies of relevant tasks.  ...  To improve transfer efficiency and guarantee optimality, we develop a novel policy reuse method, called Context-Aware Policy reuSe (CAPS), that enables multi-policy transfer.  ...  While OC learns multiple source policies in the form of options from scratch, CAPS transfers the learned policies efficiently to a new task.  ... 
arXiv:1806.03793v4 fatcat:v7enulc6bndlxjqdz4bnjnwhee
« Previous Showing results 1 — 15 out of 1,244 results