The Effective Statistician - in association with PSI The Most Important Things You Need to Know About AI And ML
Oct 21, 2024
Pietro Mascheroni, a physicist at Beringer Ingelheim specializing in machine learning and AI, dives into the transformative power of large language models (LLMs) in pharmaceutical research. He discusses their practical applications like document retrieval and data exploration. Pietro emphasizes the necessity of human oversight to ensure accuracy, while explaining how LLMs enhance team communication and collaboration. He highlights the challenges of validating AI models and the evolving capabilities of these technologies, reinforcing their potential to streamline workflows.
AI Snips
Chapters
Transcript
Episode notes
Pietro's LLM Introduction
- Pietro first encountered large language models (LLMs) in university courses and deeply engaged after ChatGPT's release.
- ChatGPT democratized access, making these models widely usable without complex setup.
Daily LLM Use Best Practices
- Use LLMs daily for various tasks, not just general queries, but embedded in tailored applications.
- Build infrastructure around models to handle real work carefully, ensuring results are relevant and accurate.
LLMs for Document and Data Workflows
- LLMs excel at navigating complex document databases and rewriting retrieved info understandably.
- They can also iteratively refine code outputs and assist in preliminary data exploration and engineering.