A quick and intuitive explanation of gradient descent

This short writeup is from Udacity’s Deep Learning Nanodegree. Journey to the Bottom of the Valley Here I’ll give you a little refresher on gradient descent so we can start training our network with MiniFlow. Remember that our goal is to make our network output as close as possible to the target values by minimizing … Read more

Cómo practicar Mindfulness al caminar

Los beneficios de la meditación ya fueron confirmados por cientos de estudios. El problema es que muchas veces es difícil empezar a meditar para personas que no conocen la práctica. Los que no conocen piensan que es necesario sentarse con piernas cruzadas al estilo monje budista. Repetidas veces me dijeron personas que para ellos, caminar … Read more

10% Happier — key concepts

Learn how to become happier and more successful through meditation After suffering a panic attack, Dan Harris began a journey to search for a solution. His search eventually led him to meditation. In this book, the author tells how meditation helped him calm down his inner narrator which was the originator of his crisis, and how … Read more

Does Online Education work? Some thoughts

I’m one of those happy guys who participated in the legendary 2011 Stanford MOOC, Introduction to Artificial Intelligence. That course made me (and thousands of students around the world) discover a whole different and revolutionary method of learning. I think today there is no question that Online Education works, but sometimes the question still comes … Read more

Linear regression with gradient descent

This post is based on the linear regression live coding session from Siraj Raval (Udacity). What is Linear Regression? Linear regression is a widely used method to find a predictor variable which describes an outcome variable. In simpler words, it describes the relationship of two value sets. In this example, Siraj shows how to do … Read more

Aplicando Deep Learning Style transfer

En estos días empece con el curso de Deep Learning de Udacity. El primer experimento que hicimos fue transferir estilos de dibujos con Convolutional Neuronal Networks (CNN’s) basados en style-transfer. Entrare en mas detalle de como funcionan en futuros posts pero in síntesis se trata de entrenar a un modelo de redes neuronales con las obras … Read more