- wldeng·Last edited by wldeng on Mar 13, 2022Linked with GitHubContributed by
Neural network in practice
tags: machine learning
https://coursera.org/share/a1e85b2f8a59047d139b389f189f1fcc
Read more
Hydrodynamic particle removal from surfaces
[toc] Background The coronavirus SARS-CoV-2 is the cause of the COVID-19 pandemic which has emerged as a serious public health threat. The schematic of its structure is shown in the figure below^1. Its diameter is of approximately 60–140 nm[^2]. Airborne droplets from an infected person's cough, sneeze, or even talking are a major source of viral spread. These droplets land on the mucous membranes of potential hosts (mouth, noses, eyes) or on hard surfaces. There the virus will be dispersed for next infection. A (scientific) question we may ask is if the virus could be removed by mucus secreted by mucous membranes under some conditions. This problem is similar to the hydrodynamic particle removal from surfaces by a liquid flow. ABAQUS simulations
May 25, 2020Proof of back propagation algorithm
[toc] In the machine learning note, we provide a brief summary of machine learning basics and concepts. Here we present a detailed proof of back propagation algorithm for the neural network. The material is mainly borrowed from References^1. Note in the convention I use here for $\Theta^{(l)}$ is consistent with Andrew Ng's notation, which indicates the matrix of weights controlling mapping from layer $l$ to $l+1$; while in^1 it means mapping from layer $l-1$ to $l$. Conventions Let us define a neural network with the following conventions: $$
Apr 26, 2020Machine learning basics
[toc] The material is from Andrew Ng's course: Machine Learning^1a. Another excellent material is the book "Neutral networks and deep learning" written by Michael Nielsen^1b. Linear regression with one variable Training sample $(x,y)$, where $x$ is called the feature (input), $y$ the target (output). $(x^{(i)},y^{(i)})$---the $i^{th}$ instance of the training sample. Hypothesis: $h_{\theta}(x) = \theta_0 + \theta_1 x$. Given a value of the feature, predict the value of the target.
Apr 25, 2020
HackMD