We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Developing Machine Learning for Impact • Anna Via • GOTO 2023
Understand the importance of problem-solving, data quality, explainability, and teamwork in developing machine learning solutions that drive business impact, and discover the keys to successful deployment and collaboration.
- Understand the problem before starting a machine learning project: Start from the problem, not the solution. Identify the business understanding, data quality, and features to develop a successful project.
- Data quality is crucial: Without good data quality, machine learning won’t work. Ensure data is labeled, relevant, and complete.
- Explainability is important: Models need to be explainable to ensure transparency and trust. Techniques like feature attribution and model interpretability can help.
- Multidisciplinary teams are essential: Machine learning projects require collaboration between data scientists, machine learning engineers, and other stakeholders to ensure successful deployment.
- High stakes projects require careful planning: Projects with high stakes require careful planning, testing, and evaluation to ensure success.
- Experimentation is key: Experimentation and A/B testing are crucial to understand the impact of machine learning models on the business.
- The importance of deployment: Deployment is a critical step in the machine learning lifecycle. Ensure models are properly deployed and integrated into the platform.
- Maturity of machine learning platforms: Companies need to have a mature machine learning platform to ensure successful deployment of models.
- The role of ethics in machine learning: Ethics are important in machine learning. Ensure models are fair, transparent, and unbiased.
- Collaboration is key: Collaboration between data scientists, machine learning engineers, and other stakeholders is essential for successful machine learning projects.
- The importance of feedback: Feedback is crucial in machine learning projects to understand the impact of models and make improvements.