3,728 Hits in 12.9 sec

Learning and Fast Adaptation for Grid Emergency Control via Deep Meta Reinforcement Learning [article]

Renke Huang, Yujiao Chen, Tianzhixi Yin, Qiuhua Huang, Jie Tan, Wenhao Yu, Xinya Li, Ang Li, Yan Du
2022 arXiv   pre-print
Towards this end, great progress has been made in developing deep reinforcement learning (DRL) based grid control solutions in recent years.  ...  In this paper, we mitigate these limitations by developing a novel deep meta reinforcement learning (DMRL) algorithm.  ...  In this paper, we tackle this problem by developing a novel deep meta-reinforcement learning (DMRL) algorithm and applying it to learn and adapt power system emergency control policies against voltage  ... 
arXiv:2101.05317v2 fatcat:55lmzubo5fhrfjsderg3csbfae

Collective Intelligence for Deep Learning: A Survey of Recent Developments [article]

David Ha, Yujin Tang
2022 arXiv   pre-print
State-of-the-art deep learning models are known to suffer from issues that range from poor robustness, inability to adapt to novel task settings, to requiring rigid and inflexible configuration assumptions  ...  We hope this review can serve as a bridge between the complex systems and deep learning communities.  ...  Deep Reinforcement Learning The rise in deep learning gave birth to the use of deep neural networks for reinforcement learning, or deep reinforcement learning (Deep RL), equipping reinforcement learning  ... 
arXiv:2111.14377v3 fatcat:dg5uvn7mt5g5ncgtrzw3a3ul4y

Accelerated Deep Reinforcement Learning Based Load Shedding for Emergency Voltage Control [article]

Renke Huang, Yujiao Chen, Tianzhixi Yin, Xinya Li, Ang Li, Jie Tan, Wenhao Yu, Yuan Liu, Qiuhua Huang
2020 arXiv   pre-print
Deep reinforcement learning (DRL) was regarded and adopted as a promising approach for fast and adaptive grid stability control in recent years.  ...  To overcome these issues, an accelerated DRL algorithm named PARS was developed and tailored for power system voltage stability control via load shedding.  ...  This paper focuses on developing an accelerated deep reinforcement learning (DRL)-based control method to make load shedding for emergency voltage control fast, adaptive, and scalable.  ... 
arXiv:2006.12667v2 fatcat:vhhl5ipzwvbfthk4regiolu63y

Sparse Meta Networks for Sequential Adaptation and its Application to Adaptive Language Modelling [article]

Tsendsuren Munkhdalai
2020 arXiv   pre-print
We introduce Sparse Meta Networks -- a meta-learning approach to learn online sequential adaptation algorithms for deep neural networks, by using deep neural networks.  ...  The fast-weights are generated sparsely at each time step and accumulated incrementally through time providing a useful inductive bias for online continual adaptation.  ...  Meta-learning (i.e. learning-tolearn) [57, 43, 6] has emerged as a promising technique for fast training of deep neural networks by acquiring and transferring knowledge across different tasks through  ... 
arXiv:2009.01803v1 fatcat:52li37mdwvajlfmyc2hvuhcub4

A Review of Meta-Reinforcement Learning for Deep Neural Networks Architecture Search [article]

Yesmina Jaafra, Jean Luc Laurent, Aline Deruyver, Mohamed Saber Naceur
2018 arXiv   pre-print
Deep Neural networks are efficient and flexible models that perform well for a variety of tasks such as image, speech recognition and natural language understanding.  ...  Model configuration topic has been extensively studied in machine learning without leading to a standard automatic method.  ...  Neural Architecture Search Various strategies have been developed to operate CNN architectures design for the majority of which reinforcement learning has been selected as meta-controller.  ... 
arXiv:1812.07995v1 fatcat:352eyqnvqffbbbvk4fci2k2g2q

The Sensory Neuron as a Transformer: Permutation-Invariant Neural Networks for Reinforcement Learning [article]

Yujin Tang, David Ha
2021 arXiv   pre-print
We show that these sensory networks can be trained to integrate information received locally, and through communication via an attention mechanism, can collectively produce a globally coherent policy.  ...  Interactive demo and videos of our results:  ...  Rl2: Fast reinforce- ment learning via slow reinforcement learning. arXiv preprint arXiv:1611.02779, 2016. 14 [23] R. Dubey, P. Agrawal, D.  ... 
arXiv:2109.02869v2 fatcat:uf4nviubnffz3f2qqui24zc3ei

Deep Reinforcement Learning, a textbook [article]

Aske Plaat
2022 arXiv   pre-print
Developments go quickly, and we also cover advanced topics: deep multi-agent reinforcement learning, deep hierarchical reinforcement learning, and deep meta learning.  ...  The book is written for graduate students of artificial intelligence, and for researchers and practitioners who wish to better understand deep reinforcement learning methods and their challenges.  ...  MAML nds parameters 𝜃 that are easy and fast to netune, allowing the adaptation to happen in an embedding space that is well suited for Algorithm 9.1 MAML for Reinforcement Learning [243] Require: 𝑝  ... 
arXiv:2201.02135v2 fatcat:3icsopexerfzxa3eblpu5oal64

Meta-Learning in Neural Networks: A Survey [article]

Timothy Hospedales, Antreas Antoniou, Paul Micaelli, Amos Storkey
2020 arXiv   pre-print
We survey promising applications and successes of meta-learning such as few-shot learning and reinforcement learning.  ...  Finally, we discuss outstanding challenges and promising areas for future research.  ...  Meta Reinforcement Learning and Robotics Reinforcement learning is typically concerned with learning control policies that enable an agent to obtain high reward after performing a sequential action task  ... 
arXiv:2004.05439v2 fatcat:3r23tsxxkfbgzamow5miglkrye

Unit commitment considering multiple charging and discharging scenarios of plug-in electric vehicles

Zhile Yang, Kang Li, Qun Niu, Aoife Foley
2015 2015 International Joint Conference on Neural Networks (IJCNN)  
a Motor Control Task by Reinforcement Learning and Structural Synaptic Plasticity [#15358] Spueler Martin, Nagel Sebastian and Rosenstiel Wolfgang 5:40PM Adaptive-Critic-Based Control of a Synchronous  ...  Now the field has grown significantly in both depth and breadth. It has served as a breeding ground for many fast-developing areas, including deep learning and big data analytics.  ... 
doi:10.1109/ijcnn.2015.7280446 dblp:conf/ijcnn/YangLNF15 fatcat:6xlakikcfzfyhhm2spooe2j7ra

Meta-Learning in Neural Networks: A Survey

Timothy M Hospedales, Antreas Antoniou, Paul Micaelli, Amos J. Storkey
2021 IEEE Transactions on Pattern Analysis and Machine Intelligence  
We survey promising applications and successes of meta-learning including few-shot learning, reinforcement learning and architecture search.  ...  Finally, we discuss outstanding challenges and promising areas for future research.  ...  Meta Reinforcement Learning and Robotics Reinforcement learning is typically concerned with learning control policies that enable an agent to obtain high reward after performing a sequential action task  ... 
doi:10.1109/tpami.2021.3079209 pmid:33974543 fatcat:wkzeodki4fbcnjlcczn4mr6kry

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.  ...  The authors propose policy-space response oracle (PSRO), and its approximation, deep cognitive hierarchies (DCH), to compute best responses to a mixture of policies using deep RL, and to compute new meta-strategy  ... 
arXiv:1810.06339v1 fatcat:kp7atz5pdbeqta352e6b3nmuhy

Deep Reinforcement Learning Versus Evolution Strategies: A Comparative Survey [article]

Amjad Yousef Majid, Serge Saaybi, Tomas van Rietbergen, Vincent Francois-Lavet, R Venkatesha Prasad, Chris Verhoeven
2021 arXiv   pre-print
Deep Reinforcement Learning (DRL) and Evolution Strategies (ESs) have surpassed human-level control in many sequential decision-making problems, yet many open challenges still exist.  ...  After presenting their fundamental concepts and algorithms, a comparison is provided on key aspects such as scalability, exploration, adaptation to dynamic environments, and multi-agent learning.  ...  for Meta Reinforcement learning Description Experiments ref.  ... 
arXiv:2110.01411v1 fatcat:nw47ududyndyljlh4nx2gm73jq

6G Wireless Systems: A Vision, Architectural Elements, and Future Directions

Latif U. Khan, Ibrar Yaqoob, Muhammad Imran, Zhu Han, Choong Seon Hong
2020 IEEE Access  
We propose practical guidelines including deep Q-learning and federated learning-based transceivers, blockchain-based secure business models, homomorphic encryption, and distributed-ledgerbased authentication  ...  Finally, we outline and recommend several future directions. INDEX TERMS 6G, 5G, Internet of Things, Internet of Everything, federated learning, meta learning, blockchain.  ...  Additionally, the NYU WIRELESS research center, which comprises nearly 100 faculty members and graduate students, is working on communication foundations, machine learning, quantum nanodevices, and 6G  ... 
doi:10.1109/access.2020.3015289 fatcat:x5bpksqgezgg7jylzmvlipcde4

Empowering Things with Intelligence: A Survey of the Progress, Challenges, and Opportunities in Artificial Intelligence of Things [article]

Jing Zhang, Dacheng Tao
2020 arXiv   pre-print
In the Internet of Things (IoT) era, billions of sensors and devices collect and process data from the environment, transmit them to cloud centers, and receive feedback via the internet for connectivity  ...  Then, we present progress in AI research for IoT from four perspectives: perceiving, learning, reasoning, and behaving.  ...  Deep reinforcement learning is used for obtaining the optimal control policy on the illumination, nutrition, and ventilation conditions in the greenhouse.  ... 
arXiv:2011.08612v1 fatcat:dflut2wdrjb4xojll34c7daol4

Neuro-algorithmic Policies enable Fast Combinatorial Generalization [article]

Marin Vlastelica, Michal Rolínek, Georg Martius
2021 arXiv   pre-print
Although model-based and model-free approaches to learning the control of systems have achieved impressive results on standard benchmarks, generalization to task variations is still lacking.  ...  Many control problems require long-term planning that is hard to solve generically with neural networks alone.  ...  Meta-DrAC meta-learns the weights of a convolutional neural network used for data augmentation. RL2-DrAC meta-learns a policy that selects an augmentation from a pre-defined set of augmentations.  ... 
arXiv:2102.07456v1 fatcat:xrfilsln2rbytlkfhkelrc2mbi
« Previous Showing results 1 — 15 out of 3,728 results