How Machine Learning Fails with Megan Robertson

About this episode

Richard Campbell discusses the pitfalls of machine learning with guest Megan Robertson.

What can go wrong with machine learning? While at NDC in Toronto, Richard chatted with Megan Robertson about her experience with machine learning projects, often using retail datasets, and where they can go wrong. Megan talks about getting clear expectations and metrics for projects, so you know when you succeed, but then digs into the specifics of problems in machine learning, such as overfitting on test data. Your results are only as good as the data you put in, so a lot of focus goes into building good sets, carefully developing the model with those sets, and using techniques like cross-validation to ensure the model is behaving appropriately. There's a lot that can go wrong, but the results with an effective model can be very powerful - it is worth the effort!

People in this episode

Host: Richard Campbell

Guest: Megan Robertson

Topics covered

• machine learning
• data science
• retail datasets
• overfitting
• model development
• cross-validation

Keywords

• machine learning
• overfitting
• data quality
• model validation
• retail datasets
• cross-validation
• project metrics

Mentioned in this episode

Organizations: NDC

Places: Toronto

More episodes of RunAs Radio

• Data API Builder and SQL MCP with Jerry Nixon · June 3, 2026 · 37 min
• Team Productivity using Loop with Karinne Bessette · May 27, 2026 · 36 min
• UEFI Secure Boot with Richard Hicks · May 20, 2026 · 38 min
• Production LLMs with Vaishnavi Gudur · May 13, 2026 · 35 min
• Securing Active Directory with Spencer Alessi · May 6, 2026 · 37 min
• M365 Copilot vs Claude Cowork with Sharon Weaver · April 29, 2026 · 39 min