Search

Your search keyword '"iterative learning"' showing total 360 results
Start Over Descriptor "iterative learning"
360 results on '"iterative learning"'

351. Anotátor datové sady pro trénování neuronových sítí

Author

Beran, Vítězslav, Materna, Zdeněk, Schneider, Martin, Beran, Vítězslav, Materna, Zdeněk, and Schneider, Martin

Abstract

Cílem práce je návrh a implementace systému umožňujícího anotovat obrazová data pro učení neuronových sítí. Důležitou vlastností tohoto systému je iterativní provádění učení, kdy anotovanou množinu rozšiřuje i sama neuronová síť a uživatelé provádí korekce nad rozšiřujícími daty. Systém se skládá z informačního systému s vloženým nástrojem pro anotování komunikujícím s programem řídícím neuronové sítě. Informační systém je navržen podle architektonického modelu MVC. Frontend aplikace je postaven na knihovně React jazyka JavaScript. Design frontendu je proveden dle konvencí Material Design s využitím Material-UI frameworku. Frontend udržuje trvalé spojení s Python backendem s využitím protokolu WebSocket, jehož využívá i backend pro komunikaci s programem pro řízení neuronových sítí. NoSQL databázový systém využívá technologii MongoDB., The goal of this thesis is design and implementation of a system for image data annotation which will learn neural networks. An important feature of this system is the iterative approach of learning, where the annotated set is extended by the neural network itself and users make corrections of these extensions. The system consists of an information system with an embedded annotation tool. Information system communicates with neural network management program. The information system is designed according to the MVC architectural model. The frontend of the application is built on JavaScript library - React and is designed according to Material Design conventions using the Material-UI framework. The frontend maintains a permanent connection to the Python backend using the WebSocket protocol, which is also used by the backend to communicate with the neural network management program. The NoSQL database system uses MongoDB technology.

352. Anotátor datové sady pro trénování neuronových sítí

Author

Beran, Vítězslav, Materna, Zdeněk, Schneider, Martin, Beran, Vítězslav, Materna, Zdeněk, and Schneider, Martin

Abstract

Cílem práce je návrh a implementace systému umožňujícího anotovat obrazová data pro učení neuronových sítí. Důležitou vlastností tohoto systému je iterativní provádění učení, kdy anotovanou množinu rozšiřuje i sama neuronová síť a uživatelé provádí korekce nad rozšiřujícími daty. Systém se skládá z informačního systému s vloženým nástrojem pro anotování komunikujícím s programem řídícím neuronové sítě. Informační systém je navržen podle architektonického modelu MVC. Frontend aplikace je postaven na knihovně React jazyka JavaScript. Design frontendu je proveden dle konvencí Material Design s využitím Material-UI frameworku. Frontend udržuje trvalé spojení s Python backendem s využitím protokolu WebSocket, jehož využívá i backend pro komunikaci s programem pro řízení neuronových sítí. NoSQL databázový systém využívá technologii MongoDB., The goal of this thesis is design and implementation of a system for image data annotation which will learn neural networks. An important feature of this system is the iterative approach of learning, where the annotated set is extended by the neural network itself and users make corrections of these extensions. The system consists of an information system with an embedded annotation tool. Information system communicates with neural network management program. The information system is designed according to the MVC architectural model. The frontend of the application is built on JavaScript library - React and is designed according to Material Design conventions using the Material-UI framework. The frontend maintains a permanent connection to the Python backend using the WebSocket protocol, which is also used by the backend to communicate with the neural network management program. The NoSQL database system uses MongoDB technology.

353. Anotátor datové sady pro trénování neuronových sítí

Author

Beran, Vítězslav, Materna, Zdeněk, Beran, Vítězslav, and Materna, Zdeněk

Abstract

Cílem práce je návrh a implementace systému umožňujícího anotovat obrazová data pro učení neuronových sítí. Důležitou vlastností tohoto systému je iterativní provádění učení, kdy anotovanou množinu rozšiřuje i sama neuronová síť a uživatelé provádí korekce nad rozšiřujícími daty. Systém se skládá z informačního systému s vloženým nástrojem pro anotování komunikujícím s programem řídícím neuronové sítě. Informační systém je navržen podle architektonického modelu MVC. Frontend aplikace je postaven na knihovně React jazyka JavaScript. Design frontendu je proveden dle konvencí Material Design s využitím Material-UI frameworku. Frontend udržuje trvalé spojení s Python backendem s využitím protokolu WebSocket, jehož využívá i backend pro komunikaci s programem pro řízení neuronových sítí. NoSQL databázový systém využívá technologii MongoDB., The goal of this thesis is design and implementation of a system for image data annotation which will learn neural networks. An important feature of this system is the iterative approach of learning, where the annotated set is extended by the neural network itself and users make corrections of these extensions. The system consists of an information system with an embedded annotation tool. Information system communicates with neural network management program. The information system is designed according to the MVC architectural model. The frontend of the application is built on JavaScript library - React and is designed according to Material Design conventions using the Material-UI framework. The frontend maintains a permanent connection to the Python backend using the WebSocket protocol, which is also used by the backend to communicate with the neural network management program. The NoSQL database system uses MongoDB technology.

354. Anotátor datové sady pro trénování neuronových sítí

Author

Beran, Vítězslav, Materna, Zdeněk, Beran, Vítězslav, and Materna, Zdeněk

Abstract

Cílem práce je návrh a implementace systému umožňujícího anotovat obrazová data pro učení neuronových sítí. Důležitou vlastností tohoto systému je iterativní provádění učení, kdy anotovanou množinu rozšiřuje i sama neuronová síť a uživatelé provádí korekce nad rozšiřujícími daty. Systém se skládá z informačního systému s vloženým nástrojem pro anotování komunikujícím s programem řídícím neuronové sítě. Informační systém je navržen podle architektonického modelu MVC. Frontend aplikace je postaven na knihovně React jazyka JavaScript. Design frontendu je proveden dle konvencí Material Design s využitím Material-UI frameworku. Frontend udržuje trvalé spojení s Python backendem s využitím protokolu WebSocket, jehož využívá i backend pro komunikaci s programem pro řízení neuronových sítí. NoSQL databázový systém využívá technologii MongoDB., The goal of this thesis is design and implementation of a system for image data annotation which will learn neural networks. An important feature of this system is the iterative approach of learning, where the annotated set is extended by the neural network itself and users make corrections of these extensions. The system consists of an information system with an embedded annotation tool. Information system communicates with neural network management program. The information system is designed according to the MVC architectural model. The frontend of the application is built on JavaScript library - React and is designed according to Material Design conventions using the Material-UI framework. The frontend maintains a permanent connection to the Python backend using the WebSocket protocol, which is also used by the backend to communicate with the neural network management program. The NoSQL database system uses MongoDB technology.

355. Anotátor datové sady pro trénování neuronových sítí

Author

Beran, Vítězslav, Materna, Zdeněk, Schneider, Martin, Beran, Vítězslav, Materna, Zdeněk, and Schneider, Martin

Abstract

Cílem práce je návrh a implementace systému umožňujícího anotovat obrazová data pro učení neuronových sítí. Důležitou vlastností tohoto systému je iterativní provádění učení, kdy anotovanou množinu rozšiřuje i sama neuronová síť a uživatelé provádí korekce nad rozšiřujícími daty. Systém se skládá z informačního systému s vloženým nástrojem pro anotování komunikujícím s programem řídícím neuronové sítě. Informační systém je navržen podle architektonického modelu MVC. Frontend aplikace je postaven na knihovně React jazyka JavaScript. Design frontendu je proveden dle konvencí Material Design s využitím Material-UI frameworku. Frontend udržuje trvalé spojení s Python backendem s využitím protokolu WebSocket, jehož využívá i backend pro komunikaci s programem pro řízení neuronových sítí. NoSQL databázový systém využívá technologii MongoDB., The goal of this thesis is design and implementation of a system for image data annotation which will learn neural networks. An important feature of this system is the iterative approach of learning, where the annotated set is extended by the neural network itself and users make corrections of these extensions. The system consists of an information system with an embedded annotation tool. Information system communicates with neural network management program. The information system is designed according to the MVC architectural model. The frontend of the application is built on JavaScript library - React and is designed according to Material Design conventions using the Material-UI framework. The frontend maintains a permanent connection to the Python backend using the WebSocket protocol, which is also used by the backend to communicate with the neural network management program. The NoSQL database system uses MongoDB technology.

356. Iterative Learning and Denoising in Convolutional Neural Associative Memories

Author

A. Karbasi, H. Salavati, and A. Shokrollahi

Subjects
Iterative learning, Message passing, algoweb_bio, Associative memory, Stochastic learning, Convolutional neural networks, Convolutional codes, Coding theory, Neural networks
Abstract

The task of a neural associative memory is to retrieve a set of previously memorized pat- terns from their noisy versions by using a net- work of neurons. Hence, an ideal network should be able to 1) gradually learn a set of patterns, 2) retrieve the correct pattern from noisy queries and 3) maximize the number of memorized patterns while maintaining the reliability in responding to queries. We show that by considering the inherent redundancy in the memorized patterns, one can obtain all the mentioned properties at once. This is in sharp contrast with previous work that could only improve one or two aspects at the expense of the others. More specifically, we devise an iterative algorithm that learns the redundancy among the patterns. The resulting network has a retrieval capacity that is exponential in the size of the network. Lastly, by considering the local structures of the net- work, the asymptotic error correction performance can be made linear in the size of the network.

357. Exploration of Hyperdimensional Computing Strategies for Enhanced Learning on Epileptic Seizure Detection

Author

Una Pale, Tomas Teijeiro, and David Atienza

Subjects
Signal Processing (eess.SP), FOS: Computer and information sciences, hyperdimensional computing, Computer Science - Machine Learning, online learning, Computer Science - Neural and Evolutionary Computing, multi-centroid learning, Electroencephalography, iterative learning, Machine Learning (cs.LG), Machine Learning, ComputingMethodologies_PATTERNRECOGNITION, epileptic seizure detection, Seizures, FOS: Electrical engineering, electronic engineering, information engineering, Humans, epilepsy, Neural and Evolutionary Computing (cs.NE), Electrical Engineering and Systems Science - Signal Processing, Algorithms
Abstract

Wearable and unobtrusive monitoring and prediction of epileptic seizures has the potential to significantly increase the life quality of patients, but is still an unreached goal due to challenges of real-time detection and wearable devices design. Hyperdimensional (HD) computing has evolved in recent years as a new promising machine learning approach, especially when talking about wearable applications. But in the case of epilepsy detection, standard HD computing is not performing at the level of other state-of-the-art algorithms. This could be due to the inherent complexity of the seizures and their signatures in different biosignals, such as the electroencephalogram (EEG), the highly personalized nature, and the disbalance of seizure and non-seizure instances. In the literature, different strategies for improved learning of HD computing have been proposed, such as iterative (multi-pass) learning, multi-centroid learning and learning with sample weight ("OnlineHD"). Yet, most of them have not been tested on the challenging task of epileptic seizure detection, and it stays unclear whether they can increase the HD computing performance to the level of the current state-of-the-art algorithms for wearable devices, such as random forests. Thus, in this paper, we implement different learning strategies and assess their performance on an individual basis, or in combination, regarding detection performance and memory and computational requirements. Results show that the best-performing algorithm, which is a combination of multi-centroid and multi-pass, can indeed reach the performance of the random forest model on a highly unbalanced dataset imitating a real-life epileptic seizure detection application.

358. Iterative Learning of Speech Recognition Models for Air Traffic Control

Author

Hartmut Helmke, Mittul Singh, Heiko Ehr, Petr Motlicek, Ajay Srinivasamurthy, Matthias Kleinert, and Youssef Oualil

Subjects
semi-supervised learning, Iterative learning, Computer science, Speech recognition, Iterative learning control, Acoustic model, Workload, ComputerApplications_COMPUTERSINOTHERSYSTEMS, Air traffic control, law.invention, law, Operational efficiency, Semisupervised learning, Language model, Radar, Adaptation (computer science)
Abstract

Automatic Speech Recognition (ASR) has recently proved to be a useful tool to reduce the workload of air traffic controllers leading to significant gains in operational efficiency. Air Traffic Control (ATC) systems in operation rooms around the world generate large amounts of untranscribed speech and radar data each day, which can be utilized to build and improve ASR models. In this paper, we propose an iterative approach that utilizes increasing amounts of untranscribed data to incrementally build the necessary ASR models for an ATC operational area. Our approach uses a semi-supervised learning framework to combine speech and radar data to iteratively update the acoustic model, language model and command prediction model (i.e. prediction of possible commands from radar data for a given air traffic situation) of an ASR system. Starting with seed models built with a limited amount of manually transcribed data, we simulate an operational scenario to adapt and improve the models through semi-supervised learning. Experiments on two independent ATC areas (Vienna and Prague) demonstrate the utility of our proposed methodology that can scale to operational environments with minimal manual effort for learning and adaptation.

359. On the data consumption benefits of accepting increased uncertainty

Author

Eric Martin, Frank Stephan, and Arun Sharma

Subjects
Iterative learning, General Computer Science, Computer science, business.industry, Learnability, Mind change bounds, Iterative learning control, Inductive inference, Memory limitations, Context (language use), DUAL (cognitive architecture), Inductive reasoning, Memorization, Theoretical Computer Science, Identification (information), Bounded function, Artificial intelligence, business, Cognitive psychology, Computer Science(all)
Abstract

In the context of learning paradigms of identification in the limit, we address the question: why is uncertainty sometimes desirable? We use mind change bounds on the output hypotheses as a measure of uncertainty and interpret ‘desirable’ as reduction in data memorization, also defined in terms of mind change bounds. The resulting model is closely related to iterative learning with bounded mind change complexity, but the dual use of mind change bounds — for hypotheses and for data — is a key distinctive feature of our approach. We show that situations exist where the more mind changes the learner is willing to accept, the less the amount of data it needs to remember in order to converge to the correct hypothesis. We also investigate relationships between our model and learning from good examples, set-driven, monotonic and strong-monotonic learners, as well as class-comprising versus class-preserving learnability.

Catalog

Books, media, physical & digital resources
See catalog results