Trainings
Playlists
- My playlists
About Us
FAQ
Join
Log In

Advance-CTR Machine Learning for Health Seminar Series: Sarah Brown, PhD

Add to existing playlist ▼

Dr. Sarah Brown presents on “Assessing Machine Learning Problems for Fairness Before Fitting.” Dr. Brown will present her team's proposal for examining problem formulations before fitting a model in order to give equal weight to a broader range of bias mitigation strategies.

Introductory

Free

Michigan Institute for Clinical Health Research (MICHR)
1600 Huron Parkway, Building 400, Ann Arbor, MI 48109
[email protected]

Characteristics of a Translational Scientist

BOUNDARY CROSSER: Breaks down disciplinary silos and collaborates with others across research areas and professions to collectively advance the development of a medical intervention.

DOMAIN EXPERT: Possesses deep disciplinary knowledge and expertise within one or more domains of the translational science spectrum ranging from basic to clinical to public health research and domains in between.

TEAM PLAYER: Practices a team science approach by leveraging the strengths and expertise and valuing the contributions of all players on the translational science team.

PROCESS INNOVATOR: Seeks to better understand the scientific and operational principles underlying the translational process, and innovates to overcome bottlenecks and accelerate that process.

SKILLED COMMUNICATOR: Communicates with understanding with all stakeholders in the translational process across diverse social, cultural, economic and scientific backgrounds, including patients and community members.

SYSTEMS THINKER: Evaluates the complex external forces, interactions and relationships impacting the development of medical interventions, including patient needs and preferences, regulatory requirements, current standards of care, and market and business demands.

RIGOROUS RESEARCHER: Conducts research at the highest levels of rigor and transparency, possesses strong statistical analysis skills, and designs research projects to maximize reproducibility.