Neural Networks in R
Overview:
In this 3h tutorial, participants will be briefly introduced to (i) the idea of using neural networks to conduct classification, (ii) practical experience of creating a neural network model using available torch modules in R, (iii) the configuration of hyperparameters of neural networks and their influence on the model performance, and (iv) overfitting issue. This tutorial focuses on the basic ideas of neural networks and does not involve mathematics derivation heavily.
At the end of the workshop, participants will be able to:
- Understand the general pipeline of conducting classification, and the concept of activation function and loss function.
- Understand the concept of hyperparameters
- Understand the concept of overfitting
Prerequisites
- Understand the classification problem and the general pipeline of conducting prediction, including the training and testing procedure; e.g. from workshops Introduction to ML in R, or Fundamentals of ML in Python.
- Knowledge of the basics of regression is preferred.
- Knowledge of R and RStudio.
Date: Wednesday, 5 April 2023.
Time: 11:30 p.m. to 2:30 p.m.
Location: hybrid (in-person at Burnside Hall 1104, and online via Zoom).
Instructor: Qihuang Zhang, assistant professor of Epidemiology, Biostatistics, and Occupational Health (EBOH) at McGill University.
