Podcast
Optimize Decisions with ML: Prescriptive & Robust Optimization for Supply Chain and Pricing
Open original DataTalks.Club episode
Optimize Decisions with ML: Prescriptive & Robust Optimization for Supply Chain and Pricing
Original Episode
Use these links for the canonical episode and media sources.
- Open the original DataTalks.Club podcast page
- Watch on YouTube
- Listen on Spotify
- Listen on Apple Podcasts
Episode Overview
How do you turn machine learning predictions into better real-world decisions—especially under uncertainty in supply chains and pricing? In this episode, Dan Becker, Founder & CEO of Decision AI and former Google data scientist and Product Director at DataRobot, walks through prescriptive analytics and decision optimization for practical business impact. With a background that includes top Kaggle performance and contributions to TensorFlow and Keras, Dan explains how to formulate optimization problems, choose.
People
Use these links to connect the episode to guest notes.
Chapter Summary
Use these checkpoints to decide whether to open the source transcript.
- 0:00 - Podcast Introduction
- 1:10 - Introduction: Dan Becker and Decision Optimization Overview
- 3:00 - Gap: Machine Learning Predictions vs. Real-World Decisions
- 6:00 - Prescriptive Analytics: Role in ML Pipelines
- 9:00 - Formulating Optimization Problems: Objectives and Constraints
- 12:00 - Modeling Uncertainty: Robust and Stochastic Optimization
- 15:30 - Integrating Predictions into Optimization Models
- 18:45 - Aligning Loss Functions with Business Objectives
- 22:00 - Solvers & Tools: OR-Tools, Gurobi, Pyomo, Open-Source Options
- 25:30 - Scalability: Large-Scale Optimization and Approximation Techniques
- 28:30 - Use Case: Supply Chain Optimization and Resource Allocation
- 32:00 - Use Case: Pricing, Bidding, and Revenue Optimization
- 35:00 - Decision Constraints: Operational, Legal, and Ethical Considerations
- 38:00 - Evaluation Metrics: Measuring Real-World Impact of Decisions
- 41:00 - Deployment: Pipelines, Monitoring, and Feedback Loops
- 44:00 - Organizational Adoption: Cross-Functional Collaboration and Change Management
- 46:30 - Skillset Blend: Data Science, Operations Research, and Software Engineering
- 49:00 - Common Pitfalls: Mis-specified Objectives and Overfitting Decisions
- 51:30 - Robustness vs. Optimality: Trade-offs in Decision Optimization
- 54:00 - Future Trends: Automated Decisioning and Prescriptive Systems
- 56:30 - Resources: Learning Paths, Libraries, and Community Recommendations
- 59:30 - Practical Advice: Getting Started with Decision Optimization Projects