Artificial Intelligence for Advanced Statistical Analysis
9th AIC 2026 Workshop by Dr. Reetam Majumder
Workshop Overview
This hands-on workshop introduces junior African statisticians to the integration of AI and modern statistical analysis, with a particular focus on deep learning as a powerful tool for tackling complex data challenges. As data becomes increasingly high-dimensional and computationally intensive, traditional statistical methods are often complemented-and sometimes transformed-by AI-driven approaches.
Participants will gain practical experience in building and applying deep learning models using the R programming environment, specifically leveraging the torch ecosystem. The workshop is designed to guide attendees through the complete data analysis pipeline, including software setup, data preprocessing, model development, and evaluation.
Beyond predictive modeling, the course highlights how deep learning can be used for statistical inference, including density estimation, parameter estimation, and uncertainty quantification in complex or intractable models. Emphasis will be placed on connecting modern AI techniques with classical statistical principles, ensuring participants develop both practical skills and theoretical insight.
By the end of the workshop, participants will be equipped with the foundational knowledge and computational tools needed to incorporate AI methods into their own statistical research and applications across diverse domains such as public health, economics, and environmental studies.
All course materials including code, datasets, slides, and installation guides will be made freely available via GitHub to support continued learning.
Key Topics Covered
1. Optimization of loss functions and links to classical statistical inference
2. Numerical optimization techniques in R
3. Software setup for deep learning (torch ecosystem)
4. Fundamentals of deep learning
5. Designing custom neural network models:
5.1. Conditional mean estimation
5.2.Quantile estimation
5.3. Density estimation
6. Advanced topics:
6.1. Simulation-based inference<
6.2. Conformal prediction
6.3. Model assessment and comparison
Software Requirements
Participants are expected to install the following prior to the workshop:
1. R (version 4.4.x or higher)
2. RStudio or Positron
3. R packages:
3.1.torch
3.2.cito
3.3.SPQR
Note: The torch package must be installed first, as it is required for other dependencies. Detailed installation instructions will be provided.
Who Should Attend and How to Register
Participants, particularly junior statisticians including PhD students and early-career faculty members from African universities, are strongly encouraged to register and actively participate in this workshop.
Due to limited capacity, selection will be primarily on a first-come, first-served basis, while also considering merit and relevance to the applicant’s background.
All selected participants are required first to register for the main AIC 2026 conference. To participate in the workshop, applicants must complete the official workshop registration form.
In addition, participants are expected to attend a preparatory virtual session prior to the workshop to ensure proper installation of the required software and to review essential statistical concepts, enabling a more effective and engaging hands-on learning experience.
Registration Deadline : May 22nd 2026