Artificial neural network

**rede neural artificial**
	modelo computacional usado em aprendizagem de máquina, baseado em funções hierárquicas conectadas Vue simplifiée d'un réseau artificiel de neurones
	Carregar ficheiro
	Wikipédia
Áudio da pronúncia
Instância de	especialidade; área de estudo; inteligência artificial;
Subclasse de	rede neural; modelo discriminativo;
Vertente de	inteligência artificial; aprendizagem automática;
Composto de	neuron layer; função de perda; otimizador; neurônio artificial; neural processing unit;
Diferente de	rede neural;
Arquivo de autoridade
	Q192776; identificador GND: 4226127-2; número de controlo da Biblioteca do Congresso: sh90001937; identificador da Biblioteca Nacional da Dieta: 01165604; identificador AUT NKC: ph115443; identificador em BabelNet: 00057388n; identificador J9U da Biblioteca Nacional de Israel: 987007551192405171
	Reasonator; Scholia; Wikidocumentaries; ficheiro KML; Pesquisa retratada;

<nowiki>red neuronal artificial; Neuronowy nec; Gervitauganet; Rangkaian neural buatan; изкуствена невронна мрежа; rețea neurală artificială; اصطناعی عصبی جالکار; Tambajotra nerônina; Umelá neurónová sieť; штучна нейронна мережа; 人工神經網絡; 人工神经网络; Sunʼiy neyron tarmoqlari; artefarita neŭra reto; umělá neuronová síť; Vještačke neuronske mreže; কৃত্রিম নিউরাল নেটওয়ার্ক; нейросеть; umjetna neuronska mreža; mạng thần kinh nhân tạo; mākslīgais neironu tīkls; вештачка неурална мрежа; rede neural artificial; 人工神经网络; kunstig nevralt nettverk; nevralt nettverk; Süni Neyron Şəbəkələr; artificial neural network; شبكة عصبونية اصطناعية; 人工神經網絡; mesterséges neurális hálózat; Neurona-sare artifizial; rede neural artificial; искусственная нейронная сеть; Kapchisqa ankucha llika; künstliches neuronales Netz; líonra néarach; شبکه عصبی مصنوعی; 人工神經網絡; kunstigt neuralt netværk; ხელოვნური ნეირონული ქსელი; ニューラルネットワーク; Rete neural artificial; רשת עצבית מלאכותית; Artificiale neuronum rete; कृत्रिम तंत्रिका नेटवर्क; 人工神经网络; neuroverkko; Արհեստական Ջղաբջիջային Ցանց; செயற்கை நரம்பணுப் பிணையம்; rete neurale artificiale; tehisnärvivõrk; โครงข่ายประสาทเทียม; 人工神经网络; réseau de neurones artificiels; rede neural artificial; కృత్రిమ నాడీ జీవ కణజాల వల; вештачка невронска мрежа; Jîn-kang sîn-keng bāng-lō͘; dirbtinis neuroninis tinklas; umetna nevronska mreža; 人工神經網絡; Արհեստական նեյրոնային ցանց; 人工神經網路; Jaringan saraf tiruan; sieć neuronowa; കൃത്രിമ നാഡീവ്യൂഹം; kunstmatig neuraal netwerk; ଆର୍ଟିଫିସିଆଲ ନ୍ୟୁରାଲ ନେଟୱର୍କ; xarxa neuronal artificial; Yapay sinir ağları; artificiellt neuronnät; rede neural artificial; 인공 신경망; τεχνητό νευρωνικό δίκτυο; штучная нэўронная сетка; modelo computacional; modèle de calcul inspiré des neurones biologiques; математическая модель, построенная по принципу организации и функционирования биологических нейронных сетей; ankucha llika; Netz aus künstlichen Neuronen; mô hình toán học hay mô hình tính toán được xây dựng dựa trên các mạng neural sinh học; ierīce, kas paredzēta smadzeņu darbības modelēšanai; 模仿生物神經網路結構和功能的數學模型或計算模型; 主に機械学習で用いられる、人間の脳神経を模したモデルの一種; modelo computacional usado em aprendizagem de máquina, baseado em funções hierárquicas conectadas; kunstig etterligning av biologisk nervevev fra hjernen eller det sensoriske system; samlingsnamn på självlärande algoritmer som försöker efterlikna biologiska neuronnät; υπολογιστικό μοντέλο στην μορφή κυκλώματος διασυνδεδεμένων κόμβων (παρομοιάζοντας τους βιολογικούς νευρώνες) που χρησιμοποιείται στην επίλυση προβλημάτων στον χώρο της υπολογιστικής νοημοσύνης; מודל מתמטי חישובי; paradigma d'aprenentatge i processament automàtic; computational model used in machine learning, based on connected, hierarchical functions; koneoppimisessa käytetty laskennanmalli; modello matematico ispirato ai neuroni biologici; 기계학습과 인지과학에서 생물학의 신경망에서 영감을 얻은 알고리즘; modelo computacional usado na aprendizaxe automática, baseado en funcións xerárquicas conectadas; հաշվողական մոդել, որն օգտագործվում է մեքենայական ուսուցման մեջ՝ հիմնված միացված, հիերարխիկ ֆունկցիաների վրա; výpočetní model používaný ve strojovém učení; modelo computacional usado em aprendizagem de máquina, baseado em funções hierárquicas conectadas; Redes Neuronales; Redes de neuronas; Redes Neurales; Redes neuronales artificiales; Neural Network; red neural artificial; redes neurales artificiales; redes fondes; rede neural simulada; rede neural estática; rede neural; XN; XNA; xarxa neuronal; KAL; Artiphisiyal ankucha llika; ANN; Ankucha llika; artificial neural networks; künstliche neuronale Netzwerke; líonra néarógach; ANN; ԱՆՑ; 人工智慧网路; 類神經網路; 协同神经网络; 神经网络; 人工神经网络; neuralt netværk; ニューラル・ネットワーク; ニューロ; ニューラルネット; 人工ニューラルネットワーク; 神経細胞網; ニューラルネットワークモデル; 神経回路網; רשת קישרית; 人工神經網路; 神經網路; न्यूरल नेटवर्क; ఏ ఎన్ ఎన్; ఆర్టీఫిషియల్ న్యూరల్ నెట్వర్క్; 인공신경망; neuronová síť; neuronové sítě; rete neurale; ANN; reti neurali; réseaux de neurones; réseau de neurones; reseau de neurones; réseau neuronique; réseau de neuronnes; artificial Neural Network; réseau de neurone; reseau de neurones artificiels; réseaux neuroniques; réseaux neuronaux; réseau neuronal; réseau de neurones artificiel; neuronal network; réseau neuronal artificiel; närvivõrk; mạng neural nhân tạo; ANN; artificial neural network; ANN; mākslīgie neironu tīkli; neironu tīkli (datorzinātne); неурална мрежа; umetno živčno omrežje; umetno nevralno omrežje; Artificial neural network; 神经网络; искусственные нейронные сети; нейросеть; искусственная нейросеть; ИНС; нейронная сеть; Neural network; ข่ายงานประสาทเทียม; เครือข่ายประสาทเทียม; ANN; neural network; connectionist system; connectionist systems; deep nets; Simulated Neural Network; Static Neural Network; SNN; Synergetic Neural Network; Neural Networks, Computer; neural net; artificial neural networks; kunstig nevralt nettverk; Jaringan Syaraf Tiruan; Artificial Neural Network; Neuroniniai tinklai; rede neural; rede neuronal; rede neuronal artificial; нейронна мережа; штучні нейронні мережі; 神经网络; RNAs; RNA; rede neural; الشبكات العصبونية الاصطناعية; شبكات عصبية اصطناعية; الشبكات العصبية الإصطناعية; شبكات عصبونيه اصطناعيه; الشبكات العصبية الاصطناعية; νευρωνικό δίκτυο; شبکه‌های عصبی</nowiki>

English: An artificial neural network (ANN) is an interconnected group of artificial neurons that uses a mathematical or computational model for information processing based on a connectionist approach to computation.

Structure

Neurons in artificial neural networks are generally structured in layers, each layer holding several neurons. This structure can be quite different.

Three different structures are shown here:

Neural network with one layer (without description).
Neural network with one layer (english description).
Neural network with one layer (German description).

Neural network with two layers (english description).

Neural network with one recurrent layer (without description).
Neural network with one recurrent layer (english description).
Neural network with one recurrent layer (German description).

Threshold outsourced as a bias neuron.

Neuron

A artificial neural network is build from several Neurons. A Neuron can be drawn in the following way:

Artificial neuron (without description).
Artificial neuron (english description).
Artificial neuron (German description).
Artificial neuron (French description).
Artificial neuron (without description).

Activation functions

Neurons can have different activation functions.

Three different functions are described here:

Hard limit function

A neuron with a hard limit function

$\varphi ^{\mbox{hlim}}(v)={\begin{cases}1&{\mbox{for }}v\geq 0\\0&{\mbox{for }}v<0\end{cases}}$

Piecewise linear function

$\varphi ^{\mbox{pwl}}(v)={\begin{cases}1&{\mbox{for }}v\geq {\frac {1}{2}}\\v+{\frac {1}{2}}&{\mbox{for }}-{\frac {1}{2}}<v<{\frac {1}{2}}\\0&{\mbox{for }}v\leq -{\frac {1}{2}}\end{cases}}$

Sigmoid function

A sigmoid function is also called a McCulloch-Pitts Model. can have a variable slope parameter $a$

$\varphi _{a}^{\mbox{sig}}(v)={\frac {1}{1+\exp(-av)}}.$

Hard limit activation function.
Piecewise linear activation function.
Sigmoid function with slope $a=5$ .
Radial base activation function.
Hyperbolic tangens activation function.

Specific types

Recurrent neural networks

Recurrent neural network (RNN) and its unfold version
A diagram for a one-unit Long Short-Term Memory (LSTM)
A diagram for a one-unit Gated Recurrent Unit (GRU)

Elman Networks

Elman Networks are special artificial neural networks which have a memory and thus are able to represent time in an implicit way.

Elman Network without description.
Elman Network with english description.
Elman Network with german description.

Time Delay Neural Networks

Time Delay Neural Networks (TDNNs) are special artificial neural networks which receive input over several time steps. Time is represented in an explicit way. The image shows an two-layer TDNN with neuron activations.

TDNN without description.
TDNN with english description.
TDNN with german description.

Artificial neural network

Índice

Structure

Neuron