Deep Learning and Computer Vision in R: A Practical Introduction

Deep learning has transformed how we extract insights from images, powering applications such as cell classification, disease detection in scans, and tissue segmentation. While these techniques are increasingly relevant in modern research, many applied statisticians are unfamiliar with the core concepts and tools behind them.

This hands-on workshop introduces the fundamentals of deep learning and its applications in computer vision, with a focus on accessibility for R users. We will begin with a high-level overview of deep learning and neural networks, then build intuition for the anatomy of a neural network, covering layers, activation functions, model training, and key components for image analysis such as convolutional and pooling layers. Practical applications will include:

• Image classification (e.g. distinguishing image types or categories)
• Object detection (e.g. locating objects within images)
• Image segmentation (time permitting)

All work will be done in R using the reticulate package to access Python-based deep learning tools. To streamline the experience, participants will use a pre-configured Posit Cloud workspace, accessible entirely through a web browser.

Requirements

Participants should:

• Create a free Posit Cloud account (details and project link will be provided)
• Be comfortable with basic R tasks (e.g. loading data, using functions, navigating RStudio)
• Have a working knowledge of basic statistical concepts (e.g. regression, classification, model accuracy)

Desirable

• No prior experience with deep learning, computer vision, or Python is required
• Some familiarity with matrices or linear algebra may help conceptual understanding, but is not necessary

Biography

Dr Patrick (Weihao) Li is a Postdoctoral Research Fellow at the Australian National University, where his research centers on applying machine learning and computer vision techniques to the grains industry. For his PhD, he developed computer vision models to automate the assessment of residual plots in regression diagnostics. An award-winning software developer, Patrick has published multiple R packages on CRAN and GitHub, contributing to the open-source ecosystem for data analysis and visualization.

Cancellations policy

a. Registration cancellations must be made in writing to the Chair of the Local Organising Committee (emi.tanaka@anu.edu.au) for consideration by the committee. 

b. Cancellations after November 8, 2025 may incur some penalty.