Skip to content
    About

    Categories

    • Artificial Intelligence 168
    • Technology 144
    • Architecture 91
    • Methodologies 75
    • Tools 55
    • Software Development 54
    • Industry 4.0 25
    • User Experience 21
    • How-To 19
    • Startup 15

    Tag

    deep-learning

    Filter #algebra-lineal 1 #aprendizaje-automatico 1 #aprendizaje-profundo 1 #autoaprendizaje 1 #backpropagation 1 #bert 1 #clasificacion 1 #clasificacion-binaria 1 #deep-learning 1 #dinov2 1 #fine-tuning 1 #formulacion-matematica 1
    DINOv2: Advances in Self-Supervised Computer Vision
    Artificial Intelligence

    DINOv2: Advances in Self-Supervised Computer Vision

    DINOv2 de Meta AI entrena modelos de visión por computadora sin etiquetas humanas, con resultados que superan a modelos supervisados en tareas de clasificación, segmentación y profundidad.

    76 8 min April 24, 2023 4.5
    The Hyperbolic Tangent: A Powerful Activation Function
    Artificial Intelligence

    The Hyperbolic Tangent: A Powerful Activation Function

    La tangente hiperbólica (tanh) produce salidas simétricas entre -1 y 1, lo que la convierte en una función de activación más estable que la sigmoide para capas ocultas.

    138 7 min March 24, 2023 4.3
    The Sigmoid Function: A Key Tool in Neural Networks
    Artificial Intelligence

    The Sigmoid Function: A Key Tool in Neural Networks

    La función sigmoide comprime cualquier valor de entrada en el rango (0, 1), lo que la convierte en la función de activación natural para modelar probabilidades en redes neuronales.

    282 6 min March 24, 2023 4.2
    Softmax Function: Activation for Classification
    Artificial Intelligence

    Softmax Function: Activation for Classification

    La función Softmax convierte vectores de salida de una red neuronal en distribuciones de probabilidad. Es el estándar para clasificación multiclase y el fundamento matemático de los modelos de lenguaje.

    177 9 min March 24, 2023 4.2
    The Leaky ReLU Function and Its Role in Neural Networks
    Artificial Intelligence

    The Leaky ReLU Function and Its Role in Neural Networks

    Leaky ReLU resuelve el problema de la neurona muerta de ReLU estándar al permitir un gradiente pequeño en la región negativa, mejorando el entrenamiento en redes profundas.

    142 6 min March 24, 2023
    The Rectified Linear Unit (ReLU): An Essential Tool for Deep Learning
    Artificial Intelligence

    The Rectified Linear Unit (ReLU): An Essential Tool for Deep Learning

    ReLU es la función de activación más utilizada en redes neuronales profundas: simple, eficiente y resistente al desvanecimiento del gradiente que lastra a la sigmoide.

    86 6 min March 24, 2023
    The Step Function: An Essential Tool in Neural Networks
    Artificial Intelligence

    The Step Function: An Essential Tool in Neural Networks

    La función escalón o de Heaviside es la función de activación más simple de una red neuronal: convierte cualquier entrada en una salida binaria 0 o 1.

    116 6 min March 24, 2023 4.6
    Linear Function: A Common Activation Function
    Artificial Intelligence

    Linear Function: A Common Activation Function

    La función lineal es la función de activación más simple en redes neuronales. Útil para regresión, pero con limitaciones críticas para capas ocultas: no introduce no linealidad.

    49 9 min March 24, 2023 4.4
    The Fully Connected Neural Network: An Innovative Approach to Machine Learning
    Artificial Intelligence

    The Fully Connected Neural Network: An Innovative Approach to Machine Learning

    La red neuronal totalmente conectada o densa es el bloque fundamental del aprendizaje profundo: cada neurona se conecta con todas las de la capa anterior y posterior.

    95 7 min March 24, 2023 4.4
    Mathematical Formulation of Artificial Neural Network Input
    Artificial Intelligence

    Mathematical Formulation of Artificial Neural Network Input

    Cómo se representan matemáticamente las entradas, pesos y funciones de activación en una red neuronal artificial, y cómo el algoritmo de retropropagación ajusta esos pesos durante el entrenamiento.

    59 9 min March 24, 2023 4.3
    Multilayer Neural Networks: Advancing Artificial Intelligence
    Artificial Intelligence

    Multilayer Neural Networks: Advancing Artificial Intelligence

    Cómo funcionan las redes neuronales multicapa, qué las hace tan poderosas y por qué el deep learning depende de ellas para resolver problemas complejos.

    50 9 min March 24, 2023 4.5
    Pre-trained Models and Transfer Learning
    Artificial Intelligence

    Pre-trained Models and Transfer Learning

    La transferencia de aprendizaje permite reutilizar modelos entrenados en grandes conjuntos de datos para resolver tareas nuevas con mucho menos datos y tiempo de cómputo. Cómo funciona y cuándo usarla.

    88 11 min March 18, 2023 4.1

    Posts pagination

    1 2 Next
    Javier Cañete Arroyo — JACAR La marca JACAR toma las dos primeras letras de cada parte del nombre: JAvier, CAñete, ARroyo. CEO · FUNDADOR JAvierCAñeteARroyo JACAR JAvier · CAñete · ARroyo

    Production notes on technology, DevOps, and AI from Madrid.

    LinkedIn RSS Jacar Systems

    Explore

    • Artificial Intelligence 168
    • Technology 144
    • Architecture 91
    • Methodologies 75
    • Tools 55
    • Software Development 54

    Latest

    • How to build a production-ready agent with the Anthropic SDK, step by step 12 May 2026
    • Claude Code vs Cursor vs GitHub Copilot in 2026: a comparison with measured tasks 8 May 2026
    • MCP (Model Context Protocol) in 2026: the complete guide for engineering teams 4 May 2026
    • Essential Software for Your New M5 Mac (2026 guide) 1 May 2026

    Connect

    Say hello hola@jacar.es
    Location
    Madrid, ES
    Languages
    ES · EN
    Archive
    438 entries
    Topics #agentes ia #llm #anthropic #ia #agentes #observabilidad #devops #kubernetes #seguridad #contenedores #ia generativa #industria 4.0 #rag #openai #inteligencia artificial #deep-learning
    © 2026 Jacar Systems S.L. · made with in · Madrid, Spain Systems operational
    Privacy Cookies RSS Search
    Manage Consent
    To provide the best experiences, we use technologies like cookies to store and/or access device information. Consenting to these technologies will allow us to process data such as browsing behavior or unique IDs on this site. Not consenting or withdrawing consent, may adversely affect certain features and functions.
    Functional Always active
    The technical storage or access is strictly necessary for the legitimate purpose of enabling the use of a specific service explicitly requested by the subscriber or user, or for the sole purpose of carrying out the transmission of a communication over an electronic communications network.
    Preferences
    The technical storage or access is necessary for the legitimate purpose of storing preferences that are not requested by the subscriber or user.
    Statistics
    The technical storage or access that is used exclusively for statistical purposes. The technical storage or access that is used exclusively for anonymous statistical purposes. Without a subpoena, voluntary compliance on the part of your Internet Service Provider, or additional records from a third party, information stored or retrieved for this purpose alone cannot usually be used to identify you.
    Marketing
    The technical storage or access is required to create user profiles to send advertising, or to track the user on a website or across several websites for similar marketing purposes.
    • Manage options
    • Manage services
    • Manage {vendor_count} vendors
    • Read more about these purposes
    View preferences
    • {title}
    • {title}
    • {title}