AI/ML in Regulated (GxP) Life Sciences Sectors - Project Phase: Iterative training and fine-tuning
Complimentary
Learning Level: Basic
Time: 1100 - 1230 EDT
Session Length: 1.5 hours
Guiding the pharmaceutical and medical device industries to successfully transition into the era of AI/ML and Pharma 4.0, the presenters will provide tangible insights derived from GAMP and quality perspectives to ensure product quality, patient safety and data integrity. We will build on the foundations presented in the first two webinars (a) concept phase and b) considerations on data sets and representativeness) by discussing the system engineering and training stages of AI/Ml subsystem development. Example studies and use cases will accompany this webinar to provide further practical insights.
Learning Objectives:
- Provide an overview of applicable types of AI/ML algorithms and models and their applicability for potential use cases or tasks. A top-level classification will include Regression, Classification, Dimension Reduction and Deep Learning (including Generative Adversarial Networks (GANs) and Generative AI).
- Provide insights into how a model, once designed, is engineered and realized on a technical level. Describe the coding process and the use of third party libraries to facilitate the task and enable data to be imputed and the outputs presented. Gain an understanding of how specialized resources such as hardware or cloud compute may be required and used.
- Understand how a machine learning model is trained and evaluated depending on the type of algorithm used. Some machine learning algorithms require only a single-shot training cycle while others require iterative training where the internal model parameters are refined over multiple iterations until a termination condition is met. Understand how to assess models and perform a comparison of performance between competing models.
