2,801 Hits in 6.1 sec

Continual Prototype Evolution: Learning Online from Non-Stationary Data Streams [article]

Matthias De Lange, Tinne Tuytelaars
2021 arXiv   pre-print
However, learning prototypes online from streaming data proves a challenging endeavor as they rapidly become outdated, caused by an ever-changing parameter space during the learning process.  ...  Additionally, continual learning does not assume the data stream to be stationary, typically resulting in catastrophic forgetting of previous knowledge.  ...  In this challenging setup, we proposed Continual Prototype Evolution (CoPE) as a prototypical solution to learn online from non-stationary data streams.  ... 
arXiv:2009.00919v4 fatcat:xcdrovmq7rgilf3hlin7j5tnqu

The ubiquitous self-organizing map for non-stationary data streams

Bruno Silva, Nuno Cavalheiro Marques
2015 Journal of Big Data  
We perform parameter sensitivity analysis and our experiments show that UbiSOM outperforms existing proposals in continuously modeling possibly non-stationary data streams, converging faster to stable  ...  models when the underlying distribution is stationary and reacting accordingly to the nature of the change in continuous real world data streams. which permits unrestricted use, distribution, and reproduction  ...  Similarly, we used the mean neuron activity [(Mean (t)] to measure in a single value the proportion of utilized neurons during learning from stationary and non-stationary data streams.  ... 
doi:10.1186/s40537-015-0033-0 fatcat:zgnbn4ky55hd7byf73dlhpmsru

Learning from non-stationary data using a growing network of prototypes

Alejandro Cervantes, Pedro Isasi, Christian Gagne, Marc Parizeau
2013 2013 IEEE Congress on Evolutionary Computation  
Learning from non-stationary data requires methods that are able to deal with a continuous stream of data instances, possibly of infinite size, where the class distributions are potentially drifting over  ...  For handling such datasets, we are proposing a new method that incrementally creates and adapts a network of prototypes for classifying complex data received in an online fashion.  ...  INTRODUCTION Learning from non-stationary data, or Non-Stationary Learning (NSL), has become a promising field specially with the increasing availability of continuous data sources from ubiquitous computing  ... 
doi:10.1109/cec.2013.6557887 dblp:conf/cec/CervantesIGP13 fatcat:lnlvklh4trhk7pqru2kej3kwdm

A Clustering-based Framework for Classifying Data Streams [article]

Xuyang Yan, Abdollah Homaifar, Mrinmoy Sarkar, Abenezer Girma, Edward Tunstel
2021 arXiv   pre-print
The non-stationary nature of data streams strongly challenges traditional machine learning techniques.  ...  from the data streams.  ...  This non-stationary nature of data streams, known as concept drift and concept evolution [Masud et al., 2010; Gama et al., 2014] , requires a continuous learning capability for traditional machine learning  ... 
arXiv:2106.11823v1 fatcat:56fcfm2ohzbwzn4pmfgomcs6bm

Prototype-Based Classifiers in the Presence of Concept Drift: A Modelling Framework [chapter]

Michael Biehl, Fthi Abadi, Christina Göpfert, Barbara Hammer
2019 Green Energy and Technology  
We present a modelling framework for the investigation of prototype-based classifiers in non-stationary environments.  ...  Specifically, we study Learning Vector Quantization (LVQ) systems trained from a stream of high-dimensional, clustered data. We consider standard winner-takes-all updates known as LVQ1.  ...  -J. de Vries for useful discussions of earlier projects on LVQ training in stationary environments.  ... 
doi:10.1007/978-3-030-19642-4_21 fatcat:ri6rmqvlijbadnyy5l5nf4zomy

Tracking the Evolution of Smartphone Sensing for Monitoring Human Movement

Michael del Rosario, Stephen Redmond, Nigel Lovell
2015 Sensors  
In this paper, an overview of the sensors that can be found in the smartphone are presented, followed by a summary of the developments in this field with an emphasis on the evolution of algorithms used  ...  These algorithms may enable clinicians to "close the loop" by prescribing timely interventions to improve or maintain wellbeing in populations who are at risk of falling or suffer from a chronic disease  ...  Sensor data stream from the micro-electro-mechanical systems (MEMS) sensors within a smartphone.  ... 
doi:10.3390/s150818901 pmid:26263998 pmcid:PMC4570352 fatcat:dw27m56y6bgebcnhjo4kal7qsi

A comprehensive survey on machine learning for networking: evolution, applications and research opportunities

Raouf Boutaba, Mohammad A. Salahuddin, Noura Limam, Sara Ayoubi, Nashid Shahriar, Felipe Estrada-Solano, Oscar M. Caicedo
2018 Journal of Internet Services and Applications  
In this way, readers will benefit from a comprehensive discussion on the different learning paradigms and ML techniques applied to fundamental problems in networking, including traffic prediction, routing  ...  Primarily, this is due to the explosion in the availability of data, significant improvements in ML techniques, and advancement in computing capabilities.  ...  [85] RL CMAC-NN (Online) Prototype Appli- cation Patterns of Ping Flood and UDP Packet Storm attacks -3 Layers NN -Prototype developed w/ C & Matlab [407] RL Q-Learning (Online)  ... 
doi:10.1186/s13174-018-0087-2 fatcat:jvwpewceevev3n4keoswqlcacu

Statistical Mechanics of On-Line Learning Under Concept Drift

Michiel Straat, Fthi Abadi, Christina Göpfert, Barbara Hammer, Michael Biehl
2018 Entropy  
In both cases, the target, i.e., the classification or regression scheme, is considered to change continuously while the system is trained from a stream of labeled data.  ...  We exemplify the approach in terms of two specific model situations: In the first, we consider the learning of a classification scheme from clustered data by means of prototype-based Learning Vector Quantization  ...  A particularly relevant case would be that of non-stationary prior probabilities p σ (α) in classification, where a varying fraction of examples represents each of the classes in the data stream.  ... 
doi:10.3390/e20100775 pmid:33265863 fatcat:qcspe6g5zrffdijdbrcmopobua

Evolution of Non-Terrestrial Networks From 5G to 6G: A Survey [article]

M. Mahdi Azari, Sourabh Solanki, Symeon Chatzinotas, Oltjon Kodheli, Hazem Sallouha, Achiel Colpaert, Jesus Fabian Mendoza Montoya, Sofie Pollin, Alireza Haqiqatnejad, Arsham Mostaani, Eva Lagunas, Bjorn Ottersten
2022 arXiv   pre-print
Our survey further includes the major progress and outcomes from academic research as well as industrial efforts representing the main industrial trends, field trials, and prototyping towards the 6G networks  ...  Non-terrestrial networks (NTNs) traditionally have certain limited applications.  ...  In online missions, on the other hand, the UAV acts as a mobile relay, forwarding data from IoT nodes to a terrestrial gateway in real-time [130] . Li et al.  ... 
arXiv:2107.06881v2 fatcat:ap7uchcpqzbchohcn3pnrlubzm

Data stream clustering

Jonathan A. Silva, Elaine R. Faria, Rodrigo C. Barros, Eduardo R. Hruschka, André C. P. L. F. de Carvalho, João Gama
2013 ACM Computing Surveys  
Nevertheless, data stream clustering imposes several challenges to be addressed, such as dealing with non-stationary, unbounded data that arrive in an online fashion.  ...  Data stream mining is an active research area that has recently emerged to discover knowledge from large amounts of continuously generated data.  ...  Ubiquitous Data Streams (ref.  ... 
doi:10.1145/2522968.2522981 fatcat:7uuyd35nuzhmdnav5knfruwo5i

Ensemble learning for data stream analysis: A survey

Bartosz Krawczyk, Leandro L. Minku, João Gama, Jerzy Stefanowski, Michał Woźniak
2017 Information Fusion  
Furthermore, due to the non-stationary characteristics of streaming data, prediction models are often also required to adapt to concept drifts.  ...  Out of several new proposed stream algorithms, ensembles play an important role, in particular for non-stationary environments.  ...  This issue was Ensemble learning from data streams Supervised learning for classification Chunk-based ensembles for stationary streams Online ensembles for stationary streams Chunk-based ensembles  ... 
doi:10.1016/j.inffus.2017.02.004 fatcat:rfc735znxjcwdebcbjxbyx7xki

State-of-the-art on clustering data streams

Mohammed Ghesmoune, Mustapha Lebbah, Hanene Azzag
2016 Big Data Analytics  
In the literature of data stream clustering methods, a large number of algorithms use a two-phase scheme which consists of an online component that processes data stream points and produces summary statistics  ...  The traditional set-up where a static dataset is available in its entirety for random access is not applicable as we do not have the entire dataset at the launch of the learning, the data continue to arrive  ...  Acknowledgements This work has been supported by the French Foundation FSN, PIA Grant Big data-Investissements d'Avenir.  ... 
doi:10.1186/s41044-016-0011-3 fatcat:dimp634rczf7don7jk4tfunvfm

A taxonomic look at instance-based stream classifiers

Iain A.D. Gunn, Álvar Arnaiz-González, Ludmila I. Kuncheva
2018 Neurocomputing  
A key challenge when learning from such streams is the problem of concept drift. Many methods, including many prototype methods, have been proposed in recent years to address this problem.  ...  A taxonomic look at instance-based stream classifier. Neurocomputing, 286, 167-178. Abstract Large numbers of data streams are today generated in many fields.  ...  This project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No 731593.  ... 
doi:10.1016/j.neucom.2018.01.062 fatcat:64gjd5kldna6vlyzkjyqjc4io4

Supervised Learning in the Presence of Concept Drift: A modelling framework [article]

Michiel Straat, Fthi Abadi, Zhuoyun Kan, Christina Göpfert, Barbara Hammer, Michael Biehl
2021 arXiv   pre-print
We investigate so-called student teacher scenarios in which the systems are trained from a stream of high-dimensional, labeled data.  ...  We present a modelling framework for the investigation of supervised learning in non-stationary environments.  ...  Similarly, in many technical contexts, training data is available as a non-stationary stream of observations.  ... 
arXiv:2005.10531v2 fatcat:2akpjvsatfbmbfyqsmonczv52q

Evolution of XR Research in Brazil according to the first 22 SVR editions

Fabiana F F Peres, João Marcelo Teixeira
2021 Journal on Interactive Systems  
In 2016, Abreu et al. (2016) evaluate the usage of electromyogram (EMG) data provided by the Myo armband as features for classification of 20 stationary letter gestures from the Brazilian Sign Language  ...  This work was a direct evolution from the previous one.  ... 
doi:10.5753/jis.2021.2088 fatcat:lxcfnnkiijespm4hrktgmthjoe
« Previous Showing results 1 — 15 out of 2,801 results