55,660 Hits in 1.8 sec

Variational online learning of neural dynamics [article]

Yuan Zhao, Il Memming Park
2020 arXiv   pre-print
We developed a flexible online learning framework for latent nonlinear state dynamics and filtered latent states.  ...  It brings the challenge of learning both latent neural state and the underlying dynamical system because neither is known for neural systems a priori.  ...  We hope that further development would enable on-the-fly analysis of high-dimensional neural spike train during electrophysiological experiments.  ... 
arXiv:1707.09049v5 fatcat:oq5wq7jwprbd5fiywiu3zbwv74

Variational Online Learning of Neural Dynamics

Yuan Zhao, Il Memming Park
2020 Frontiers in Computational Neuroscience  
We developed a flexible online learning framework for latent non-linear state dynamics and filtered latent states.  ...  This brings with it the challenge of learning both latent neural state and the underlying dynamical system because neither are known for neural systems a priori.  ...  Online adaptive learning: Our target application scenarios are streaming data. This allows the inference during an experiment or as part of a neural prosthetics.  ... 
doi:10.3389/fncom.2020.00071 pmid:33154718 pmcid:PMC7591751 fatcat:4mdkuoqbyvdtnoggn242xzh4qa


Hitesh Shah, M. Gopal
2014 IFAC Proceedings Volumes  
Evolving neuro-fuzzy systems are intended to use online learning to extract knowledge from data and perform a high-level adaptation of the network structure.  ...  We develop a dynamic evolving fuzzy neural network (DENFIS) function approximation approach to RL systems. Potential of this approach is demonstrated through a case study⎯two-link robot manipulator.  ...  In this paper, we explore the potential of an alternative dynamic evolving fuzzy-neural network (dmEFuNN) for reinforcement learning algorithms.  ... 
doi:10.3182/20140313-3-in-3024.00058 fatcat:y4ptlzdqkrfshhal7hwu4biwbu

Online Adaptive Control of Non-linear Plants Using Neural Networks with Application to Temperature Control System

R. Hedjar
2007 Journal of King Saud University: Computer and Information Sciences  
To overcome this problem, a neural inverse model is added to the control scheme and an online update of the weights is provided.  ...  Although the neural inverse model controllers have demonstrated high potential in the nonconventional branch of non-linear control, their sensitivity to parameter variations and/or parameter uncertainties  ...  Abdullah Saad Alrasheed who carried out the experimental part of this work.  ... 
doi:10.1016/s1319-1578(07)80005-x fatcat:nbxak6lw2zds7p4xwm75v7gffi

Efficient transfer learning and online adaptation with latent variable models for continuous control [article]

Christian F. Perez, Felipe Petroski Such, Theofanis Karaletsos
2018 arXiv   pre-print
Traditional model-based RL relies on hand-specified or learned models of transition dynamics of the environment.  ...  We use Online Bayesian Inference of these learned latents to rapidly adapt online to changes in environments without retaining large replay buffers of recent data.  ...  We explore 5 episodes of online learning on an environment with changing dynamics, showing that the variational approach outperforms the generalist ( Fig. 3; Top) .  ... 
arXiv:1812.03399v1 fatcat:s7uyop2brvf7peqx5uk5gsueea

Dynamic Clustering

Abdelhamid Bouchachia
2012 Evolving Systems  
This latter consists of the following papers: Paper 1 by Fan and Bouguila titled "Online Variational Learning of Finite Dirichlet Mixture Models" introduces an online variational inference algorithm for  ...  Dynamic clustering as a form of unsupervised online/ incremental machine learning considers two concepts: (1) incrementality of the learning methods to devise the clustering model and (2) self-adaptation  ...  This latter consists of the following papers: Paper 1 by Fan and Bouguila titled ''Online Variational Learning of Finite Dirichlet Mixture Models'' introduces an online variational inference algorithm  ... 
doi:10.1007/s12530-012-9062-5 fatcat:fz7vkch3szfujhtudjnknavrru

Learning speech recognition from songbirds

Izzet B Yildiz, Katharina von Kriegstein, Stefan J Kiebel
2013 BMC Neuroscience  
Friston KJ, Trujillo-Barreto N, Daunizeau J: DEM: A variational treatment of dynamic systems. Neuroimage 2008, 41(3):849-885. 7.  ...  as humans: to learn and decode complex auditory input partitioned into sequences of syllables, in an online fashion [2] .  ...  as humans: to learn and decode complex auditory input partitioned into sequences of syllables, in an online fashion [2] .  ... 
doi:10.1186/1471-2202-14-s1-p210 pmcid:PMC3704558 fatcat:6awvehircjde7cxyp4cx3aeufe

A neurologically plausible implementation of statistical inference applied to random dot motion

James Martens, Chris Eliasmith
2007 BMC Neuroscience  
and with learning done in batches instead of online [1] .  ...  Valpola H, Karhunen J: An unsupervised ensemble learning method for nonlinear dynamic state-space models. Neural  ...  and with learning done in batches instead of online [1] .  ... 
doi:10.1186/1471-2202-8-s2-p207 fatcat:3gjvnytmzbhwzhy23sfnttaa4y

Adaptive Neural Trajectory Tracking Control for Flexible-Joint Robots with Online Learning [article]

Shuyang Chen, John Wen
2020 arXiv   pre-print
Our approach uses a multi-layer neural network to approximate unknown dynamics needed for the feedforward control.  ...  The internal weights of the nonlinear basis are updated by online backpropagation to further reduce the tracking error.  ...  Here, we explore the online learning of regressor Y with the collected input/output data of the system after dynamics variation.  ... 
arXiv:2003.05426v1 fatcat:z4ebyzqnv5glxpf3m3ogrnoace

Online Learning Neural Network Control of Buck-Boost Converter

W.M. Utomo, A. Bakar, M. Ahmad, T. Taufik, R. Heriansyah
2011 2011 Eighth International Conference on Information Technology: New Generations  
This paper proposes a neural network control scheme of a DC-DC buck-boost converter using online learning method. In this technique, a back propagation algorithm is derived.  ...  Furthermore, to investigate the effectiveness of the proposed controller, some operations such as starting-up and reference voltage variations are verified.  ...  Online Learning Algorithm of BPEOC After the neural network architecture is modelled, the next stage defines the learning model to update network parameters.  ... 
doi:10.1109/itng.2011.216 dblp:conf/itng/UtomoBATH11 fatcat:g5utikaipfgl7i53enndt6r6ri

Investigating the existence of periodicity in activity of neural network by novel neural signal processing technique - quantifying induced learning in cell culture [article]

Sayan Biswas
2017 bioRxiv   pre-print
A novel integrative model of neural signal processing termed as Activity Index is applied. AI variation is plotted graphically to show the evidence in periodicity of network analysis.  ...  This sheer ability allows dynamic nature of the network for which this network are ever changing.  ...  This parameter essentially plays a critical role in analysis of a dynamic neural network.  ... 
doi:10.1101/177360 fatcat:n2xtuocuyjfb3feiia6nccy7ka

A Control Scheme for Industrial Robots Using Artificial Neural Networks

M. Dinary, Abou-Hashema M. El-Sayed, Abdel Badie Sharkawy, G. Abouelmagd
2015 International Journal of Materials Mechanics and Manufacturing  
This scheme is developed to control arm robot manipulator without calculate the model parameters or dynamics, and use the online identification instead. The scheme consists of three parts.  ...  Index Terms-Industrial robots, ANN, online identification, neural control, parametric and payload uncertainty.  ...  A layered neural network is employed to learn the inverse dynamics of the unknown dynamical plant and acts as a feed forward controller to control the plant.  ... 
doi:10.7763/ijmmm.2015.v3.171 fatcat:zredwrrhojeyrlu6btmj5z5qty

A Stable Distributed Neural Controller for Physically Coupled Networked Discrete-Time System via Online Reinforcement Learning

Jian Sun, Jie Li
2018 Complexity  
For tackling these challenges, we suggest an online distributed reinforcement learning control algorithm with the one-layer neural network for each subsystem or called agents to adapt the variation of  ...  The stability of the control system with learning algorithm is analyzed; the upper bound of the tracking error and neural network weights are also estimated.  ...  Acknowledgments This work was supported in part by the National Natural Science Foundation of China under Grant 61703347.  ... 
doi:10.1155/2018/5950678 fatcat:45austwjzrbh5iuf5szp3svfni

Autonomic Provisioning with Self-Adaptive Neural Fuzzy Control for End-to-end Delay Guarantee

Palden Lama, Xiaobo Zhou
2010 2010 IEEE International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems  
We design the neural fuzzy controller as a hybrid of control theoretical and machine learning techniques.  ...  It is capable of self-constructing its structure and adapting its parameters through fast online learning. Unlike other supervised machine learning techniques, it does not require off-line training.  ...  Online Learning of Neural Fuzzy Controller The neural fuzzy controller combines fuzzy logic's reasoning with the learning capabilities of an artificial neural network.  ... 
doi:10.1109/mascots.2010.24 dblp:conf/mascots/LamaZ10 fatcat:if5js2vm4je25a7x6g3gqve7f4

2014 Index IEEE Transactions on Neural Networks and Learning Systems Vol. 25

2014 IEEE Transactions on Neural Networks and Learning Systems  
., +, TNNLS Aug. 2014 1508-1519 Dynamic Learning From Adaptive Neural Network Control of a Class of Nonaffine Nonlinear Systems. Dai, S.  ...  Paisitkriangkrai, S., +, TNNLS Apr. 2014 764-779 Stochastic Learning via Optimizing the Variational Inequalities.  ...  The Field of Values of a Matrix and Neural Networks. Georgiou, G.M., TNNLS Sep. 2014  ... 
doi:10.1109/tnnls.2015.2396731 fatcat:ztnfcozrejhhfdwg7t2f5xlype
« Previous Showing results 1 — 15 out of 55,660 results