Kickoff Workshop for Student Competition on Food Safety AI Benchmark Datasets

Jan 9, 2026 (12:00 PM - 1:00 PM Eastern Time)

Webinar Description

This 1-hour webinar will serve as the official kickoff for the AI Benchmarking Student Competition on Predictive Food Safety Models, which has been initially accepted by IAFP as a symposium for the 2026 Annual Meeting. The session will introduce participants to the competition’s goals, timeline, and evaluation criteria, and provide an overview of the curated datasets drawn from the Cornell Food Safety ML Repository, Agroknow’s Foodakai platform, and academic computer-vision sources. Presenters will outline best practices for developing reproducible ML/DL pipelines applicable to food-safety prediction tasks, including pathogen presence modeling, food-incident forecasting, and image-based bacterial classification. The webinar is designed to help teams begin work effectively ahead of the March submission deadline and will conclude with a live Q&A. This educational, non-commercial session supports IAFP’s mission by advancing data-driven approaches to food safety.

Learning Objectives:

1. Participants will describe the objectives, timeline and submission requirements for the AI benchmarking student competition in the food-safety domain.

2. Participants will identify the curated datasets available for the competition and understand how to access and prepare them for modelling.

3. Participants will outline standard machine learning/deep learning pipeline steps suitable for predictive food-safety modeling (data preparation, feature engineering, training, validation, reproducibility).

4. Participants will evaluate use-cases demonstrating how predictive modeling (environmental pathogen detection, food-incident forecasting, image classification) can be applied in food safety.

5. Participants will formulate strategies for team organisation, code management and reproducibility best practices applicable to the competition.