Filters








19,373 Hits in 5.6 sec

Smooth Exploration for Robotic Reinforcement Learning [article]

Antonin Raffin, Jens Kober, Freek Stulp
<span title="2021-06-20">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Reinforcement learning (RL) enables robots to learn skills from interactions with the real world.  ...  Consequences of the resulting shaky behavior are poor exploration, or even damage to the robot. We address these issues by adapting state-dependent exploration (SDE) to current Deep RL algorithms.  ...  Acknowledgments The work described in this paper was partially funded by the project "Reduced Complexity Models" from the "Helmholtz-Gemeinschaft Deutscher Forschungszentren".  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2005.05719v2">arXiv:2005.05719v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/zrh4yt3edfbynecj2imvp3hqou">fatcat:zrh4yt3edfbynecj2imvp3hqou</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200514004352/https://arxiv.org/pdf/2005.05719v1.pdf" title="fulltext PDF download [not primary version]" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <span style="color: #f43e3e;">&#10033;</span> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2005.05719v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Evaluating task-agnostic exploration for fixed-batch learning of arbitrary future tasks [article]

Vibhavari Dasagi, Robert Lee, Jake Bruce, Jürgen Leitner
<span title="2019-11-20">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In this work, we evaluate popular exploration methods by generating robotics datasets for the purpose of learning to solve tasks completely offline without any further interaction in the real world.  ...  Deep reinforcement learning has been shown to solve challenging tasks where large amounts of training experience is available, usually obtained online while learning the task.  ...  A key limitation in learning robotics skills with deep reinforcement learning is the cost of gathering new experience.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1911.08666v1">arXiv:1911.08666v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/cuto5ujsdzcodfbrtsofzbzsy4">fatcat:cuto5ujsdzcodfbrtsofzbzsy4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200825005805/https://arxiv.org/pdf/1911.08666v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/ce/9b/ce9b93c52114e065a3f9175341fcc1df964d48e8.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1911.08666v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Curriculum-Based Deep Reinforcement Learning for Adaptive Robotics: A Mini-Review

Gupta Kashish, Najjaran Homayoun
<span title="2021-05-06">2021</span> <i title="VIBGYOR ePress"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/nnniysroyrbyxctqtia3ummyb4" style="color: black;">International Journal of Robotic Engineering</a> </i> &nbsp;
Recent progress in deep reinforcement learning has corroborated to its potential to train such autonomous and robust agents.  ...  At the same time, the introduction of curriculum learning has made the reinforcement learning process significantly more efficient and allowed for training on much broader tasks.  ...  Introduction In this work, we briefly introduce the fields of reinforcement and curriculum learning and highlight curriculum-based deep reinforcement learning methods and their robotics applications.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.35840/2631-5106/4131">doi:10.35840/2631-5106/4131</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/tnoa4vd4yrgnpjzesxr5a3jq2m">fatcat:tnoa4vd4yrgnpjzesxr5a3jq2m</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210521014656/https://www.vibgyorpublishers.org/content/ijre/ijre-6-031.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/16/31/163122ba4122c46cbd50f7f1ce12eee84445612a.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.35840/2631-5106/4131"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Deep Reinforcement Learning for Robotic Manipulation-The state of the art [article]

Smruti Amarjyoti
<span title="2017-01-31">2017</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
The focus of this work is to enumerate the various approaches and algorithms that center around application of reinforcement learning in robotic ma- ]]nipulation tasks.  ...  Such methods worked well with continuous state and policy space of robots but failed to come up with generalized policies.  ...  The current state of the art in deep-reinforcement learning includes the algorithms employed by google deepmind research namely DQN (Deep Q network) for discrete actions and Deep deterministic policy gradients  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1701.08878v1">arXiv:1701.08878v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/72obbq4b2rhzngawrrspirk7sq">fatcat:72obbq4b2rhzngawrrspirk7sq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200827154436/https://arxiv.org/pdf/1701.08878v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/f8/0f/f80f8b8e4fdfba5238a169ebe9efbbfddbd46e5a.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1701.08878v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Implementation of Q Learning and Deep Q Network For Controlling a Self Balancing Robot Model [article]

MD Muhaimin Rahman, SM Hasanur Rashid, M.M Hossain
<span title="2018-07-22">2018</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In this paper, the implementation of two Reinforcement learnings namely, Q Learning and Deep Q Network(DQN) on a Self Balancing Robot Gazebo model has been discussed.  ...  The goal of the experiments is to make the robot model learn the best actions for staying balanced in an environment.  ...  RELATED WORKS Lei Tai and Ming Liu [2] , had worked on Mobile Robots Exploration using CNN based reinforcement learning.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1807.08272v1">arXiv:1807.08272v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ahh5fglt75a6xhywobddxnsk5m">fatcat:ahh5fglt75a6xhywobddxnsk5m</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200829020340/https://arxiv.org/pdf/1807.08272v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/70/73/70730c2da42849b51794afbb4a7f5feb49796f3b.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1807.08272v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Deep Reinforcement Learning for Robotic Manipulation with Asynchronous Off-Policy Updates [article]

Shixiang Gu and Ethan Holly and Timothy Lillicrap and Sergey Levine
<span title="2016-11-23">2016</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
However, robotic applications of reinforcement learning often compromise the autonomy of the learning process in favor of achieving training times that are practical for real physical systems.  ...  Deep reinforcement learning alleviates this limitation by training general-purpose neural network policies, but applications of direct deep reinforcement learning algorithms have so far been restricted  ...  ACKNOWLEDGEMENTS We sincerely thank Peter Pastor, Ryan Walker, Mrinal Kalakrishnan, Ali Yahya, Vincent Vanhoucke for their assistance and advice on robot set-ups, Gabriel Dulac-Arnold and Jon Scholz for  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1610.00633v2">arXiv:1610.00633v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/i3nllxmobvhy5dmnqdzmhtelqa">fatcat:i3nllxmobvhy5dmnqdzmhtelqa</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200915185334/https://arxiv.org/pdf/1610.00633v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/af/be/afbe59950a7d452ce0a3f412ee865f1e1d94d9ef.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1610.00633v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Implementation of Q learning and deep Q network for controlling a self balancing robot model

MD Muhaimin Rahman, S. M. Hasanur Rashid, M. M. Hossain
<span title="2018-12-21">2018</span> <i title="Springer Nature"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/n3bpzphp25huzes7cjfxb4b7dy" style="color: black;">Robotics and Biomimetics</a> </i> &nbsp;
In this paper, the implementations of two reinforcement learnings namely, Q learning and deep Q network (DQN) on the Gazebo model of a self balancing robot have been discussed.  ...  The goal of the experiments is to make the robot model learn the best actions for staying balanced in an environment.  ...  Related works Lei Tai and Ming Liu [2] had worked on Mobile Robots Exploration using CNN based reinforcement learning.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1186/s40638-018-0091-9">doi:10.1186/s40638-018-0091-9</a> <a target="_blank" rel="external noopener" href="https://www.ncbi.nlm.nih.gov/pubmed/30613463">pmid:30613463</a> <a target="_blank" rel="external noopener" href="https://pubmed.ncbi.nlm.nih.gov/PMC6302870/">pmcid:PMC6302870</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/tyb3lem2dff6fjwsnifm2gqlvm">fatcat:tyb3lem2dff6fjwsnifm2gqlvm</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20190220120629/http://pdfs.semanticscholar.org/2362/0d5605f26c16412eb16405bb661d137e24ec.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/23/62/23620d5605f26c16412eb16405bb661d137e24ec.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1186/s40638-018-0091-9"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> springer.com </button> </a> <a target="_blank" rel="external noopener" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6302870" title="pubmed link"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> pubmed.gov </button> </a>

Deep-Reinforcement-Learning-Based Semantic Navigation of Mobile Robots in Dynamic Environments [article]

Linh Kästner, Cornelius Marx, Jens Lambrecht
<span title="2020-08-02">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Current navigation methods depend on preexisting static maps and are error-prone in dynamic environments.  ...  On this account, we propose a reinforcement learning based local navigation system which learns navigation behavior based solely on visual observations to cope with highly dynamic environments.  ...  Deep Reinforcement Learning for Navigation With the advent of powerful neural networks, deep reinforcement learning (DRL) mitigated the bottleneck of tedious policy acquisitions by accelerating the policy  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2008.00516v1">arXiv:2008.00516v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/gnemrlpn4zckriarj4z6bnn7zi">fatcat:gnemrlpn4zckriarj4z6bnn7zi</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200930104949/https://arxiv.org/pdf/2008.00516v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/af/cf/afcf6e66fabf3114ad69293ac9d5da766ac24197.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2008.00516v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Mobile Robot Planner with Low-cost Cameras Using Deep Reinforcement Learning [article]

Minh Q. Tran, Ngoc Q. Ly
<span title="2020-12-21">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
This study develops a robot mobility policy based on deep reinforcement learning.  ...  Since traditional methods of conventional robotic navigation depend on accurate map reproduction as well as require high-end sensors, learning-based methods are positive trends, especially deep reinforcement  ...  Therefore, for the problem of this study, we take advantage of these algorithms. B. Deep reinforcement learning in Robot Navigation Reinforcement learning is widely applied to robots in general.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2012.11160v1">arXiv:2012.11160v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/d7g35yszt5ddtawtmd3sz53m3q">fatcat:d7g35yszt5ddtawtmd3sz53m3q</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201225065148/https://arxiv.org/pdf/2012.11160v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/2d/0f/2d0fbb731a7b1d08b7ddab8d3304f42b5840bc78.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2012.11160v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Multi-Robot Path Planning Method Using Reinforcement Learning

Hyansu Bae, Gidong Kim, Jonguk Kim, Dianwei Qian, Sukgyu Lee
<span title="2019-07-29">2019</span> <i title="MDPI AG"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/smrngspzhzce7dy6ofycrfxbim" style="color: black;">Applied Sciences</a> </i> &nbsp;
In addition, the robot collaboration scheme is highly depends on the conditions of each robot, such as its position and velocity.  ...  To compensate for these shortcomings, we apply Deep q learning to strengthen the learning algorithm combined with CNN algorithm, which is needed to analyze the situation efficiently.  ...  In this case, reinforcement learning is a Deep q learning that can be used in a real mobile robot environment by sharing q parameters for each robot.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/app9153057">doi:10.3390/app9153057</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/js4y2j5szzau7ob4nhtstc5idi">fatcat:js4y2j5szzau7ob4nhtstc5idi</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200213221935/https://res.mdpi.com/d_attachment/applsci/applsci-09-03057/article_deploy/applsci-09-03057.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/be/54/be543b084b09e094d3d6793a8951045a76ca6a8b.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/app9153057"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> mdpi.com </button> </a>

A Survey on Automatic Design Methods for Swarm Robotics Systems

Alaa Iskandar, Béla Kovács
<span title="2021-12-01">2021</span> <i title="Walter de Gruyter GmbH"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/wuifdlm7gzgybcfaiqyqth4uyu" style="color: black;">Carpathian Journal of Electronic and Computer Engineering</a> </i> &nbsp;
In general, they follow two-approach evolutionary algorithms like practical swarm optimization and reinforcement learning.  ...  swarm, and explaining the methods and advantages of using deep learning to reinforcement learning.  ...  This architecture is called deep Q network DQN, as shown in fig 7. [13] , [14] , [15] . Policy π: S×A R≥0 Expected Reward ∑ , (3) DEEP REINFORCEMENT LEARNING IN SWARM ROBOTIC SYSTEMS.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.2478/cjece-2021-0006">doi:10.2478/cjece-2021-0006</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ri6uvht5ybhtfbmifo463zkkka">fatcat:ri6uvht5ybhtfbmifo463zkkka</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220130013035/https://www.sciendo.com/pdf/10.2478/cjece-2021-0006" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/c4/51/c45135f1161d383bb539eb76c456a465008a8996.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.2478/cjece-2021-0006"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> Publisher / doi.org </button> </a>

GAPLE: Generalizable Approaching Policy LEarning for Robotic Object Searching in Indoor Environment [article]

Xin Ye, Zhe Lin, Joon-Young Lee, Jianming Zhang, Shibin Zheng and Yezhou Yang
<span title="2019-03-07">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In this paper, we first argue the object searching task is environment dependent while the approaching ability is general.  ...  We study the problem of learning a generalizable action policy for an intelligent agent to actively approach an object of interest in an indoor environment solely from its visual inputs.  ...  Generalization in deep reinforcement learning.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1809.08287v2">arXiv:1809.08287v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/3up3mviorjflpmo4eogxotjfdq">fatcat:3up3mviorjflpmo4eogxotjfdq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20191020092638/https://arxiv.org/pdf/1809.08287v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/c3/ae/c3ae950ca37046ee4edfd7fa2beaf9f2ec3936ec.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1809.08287v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Evaluation of physical damage associated with action selection strategies in reinforcement learning * *I. Koryakovskiy, H. Vallery and R.Babuška were supported by the European project KOROIBOT FP7-ICT-2013-10/611909

Ivan Koryakovskiy, Heike Vallery, Robert Babuška, Wouter Caarls
<span title="">2017</span> <i title="Elsevier BV"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/4oklsgzkjvbihhegxblzc6b7re" style="color: black;">IFAC-PapersOnLine</a> </i> &nbsp;
However, reinforcement learning relies on intrinsically risky exploration, which is often damaging for physical systems.  ...  However, reinforcement learning relies on intrinsically risky exploration, which is often damaging for physical systems.  ...  The authors use a deep neural network for learning both from low-dimensional state descriptions and high-dimensional renderings of the environment.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1016/j.ifacol.2017.08.1218">doi:10.1016/j.ifacol.2017.08.1218</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/2rjprrqcpzca3cd3j2geze7f6u">fatcat:2rjprrqcpzca3cd3j2geze7f6u</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20180719092949/https://repository.tudelft.nl/islandora/object/uuid%3A8df049df-64ce-430c-b4c5-e345c961d58d/datastream/OBJ/download" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/c0/55/c055f526961a51ae1541b12c3403422f2c38b17f.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1016/j.ifacol.2017.08.1218"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> elsevier.com </button> </a>

Improved deep reinforcement learning for robotics through distribution-based experience retention

Tim de Bruin, Jens Kober, Karl Tuyls, Robert Babuska
<span title="">2016</span> <i title="IEEE"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/dmucnarmarh2fj6syg5jyqs7ny" style="color: black;">2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)</a> </i> &nbsp;
Recent years have seen a growing interest in the use of deep neural networks as function approximators in reinforcement learning.  ...  This makes it attractive in scenarios where sustained exploration is in-feasible or undesirable, such as for physical systems like robots and for life long learning.  ...  This is especially true for reinforcement learning on physical systems such as robots, as the method reduces the need for continued thorough exploration, allows for improved generalization performance,  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/iros.2016.7759581">doi:10.1109/iros.2016.7759581</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/iros/BruinKTB16.html">dblp:conf/iros/BruinKTB16</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/hvdyzkf2w5edtggx3eqxtmfgha">fatcat:hvdyzkf2w5edtggx3eqxtmfgha</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20180719075920/https://repository.tudelft.nl/islandora/object/uuid%3A1f7633c7-d884-4947-8aef-0538a6cf8f74/datastream/OBJ/download" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/33/a9/33a9a939d1a0082d8528277fc423a79a62d8169a.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/iros.2016.7759581"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> ieee.com </button> </a>

Learning of Long-Horizon Sparse-Reward Robotic Manipulator Tasks with Base Controllers [article]

Guangming Wang, Minjian Xin, Wenhua Wu, Zhe Liu, Hesheng Wang
<span title="2021-12-04">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Deep Reinforcement Learning (DRL) enables robots to perform some intelligent tasks end-to-end. However, there are still many challenges for long-horizon sparse-reward robotic manipulator tasks.  ...  Built upon Deep Deterministic Policy Gradients (DDPG), our algorithm incorporates the existing base controllers into stages of exploration, value learning, and policy update.  ...  Abbeel, vision-based deep reinforcement learning for robotic motion control,” in “Overcoming exploration in reinforcement learning with demonstra- Proceedings of the  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2011.12105v3">arXiv:2011.12105v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/qgjdebnjmvgadkzdkn4vgqbpsm">fatcat:qgjdebnjmvgadkzdkn4vgqbpsm</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211208093933/https://arxiv.org/pdf/2011.12105v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/16/b0/16b03f654b259eaeb744b185191490a953af3d5f.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2011.12105v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>
&laquo; Previous Showing results 1 &mdash; 15 out of 19,373 results