European Statistical Forum
November 14, 2023 | Munich, Germany
We are pleased to share that we will present at the European Statistical Forum. This year’s ESForum aims to discuss the future direction of clinical trial design in the era of precision medicine and to debate the role of AI/ML and RWD in the rapidly changing landscape of clinical trial designs.
Do not miss our speech, “How much Training Data is needed to Train a Learner? – A Heuristic Approach,” presented by Rajat Mukherjee, Vice President of Advanced Statistics and Data Science.
Let’s meet at ESForum and discuss how we can help you to optimize sample sizes and time to decision and de-risk studies using innovative trial designs.
About Our Speech
How much Training Data is needed to Train a Learner? – A Heuristic Approach
15:00 h| November 14th
One of the most crucial phases of a biomarker discovery or diagnostics development using machine learning (ML) is training the learner algorithm.
In this talk, we focus on learners where a single dominating predictor can be identified. We propose and present a simulation-based approach to estimating the size of the training data.
We also present a seamless adaptive design where the training and validation of a machine learning-based diagnostic device can be carried out in an operationally seamless fashion while mitigating several risk factors that arise naturally in these kinds of biomedical problems.