Learning Bayesian Statistics

Bitesize | How To Model Risk Aversion In Pricing?

Today's clip is from Episode 152 of the podcast, with Daniel Saunders. In this conversation, Daniel Saunders explains how to incorporate risk aversion into Bayesian price optimization. The key insight is that uncertainty around expected profit is asymmetric across price points, low prices yield more predictable (if modest) returns, while high prices introduce much wider uncertainty. Rather than simply maximizing expected profit, you can pass profit through an exponential utility function that models diminishing returns, a well-established idea from economics. This adds an adjustable risk aversion parameter to the optimization: as risk aversion increases, the model shifts toward more conservative price recommendations, trading off potentially large but uncertain gains for outcomes with tighter, more reliable distributions.Get the full discussion here• Join this channel to get access to perks:https://www.patreon.com/c/learnbayesstats• Intro to Bayes Course (first 2 lessons free): https://topmate.io/alex_andorra/503302• Advanced Regression Course (first 2 lessons free): https://topmate.io/alex_andorra/1011122Our theme music is « Good Bayesian », by Baba Brinkman (feat MC Lars and Mega Ran). Check out his awesome work at https://bababrinkman.com/ !

#152 A Bayesian decision theory workflow, with Daniel Saunders

• Support & get perks!• Proudly sponsored by PyMC Labs! Get in touch at alex.andorra@pymc-labs.com• Intro to Bayes and Advanced Regression courses (first 2 lessons free)Our theme music is « Good Bayesian », by Baba Brinkman (feat MC Lars and Mega Ran). Check out his awesome work !Chapters:00:00 The Importance of Decision-Making in Data Science06:41 From Philosophy to Bayesian Statistics14:57 The Role of Soft Skills in Data Science18:19 Understanding Decision Theory Workflows22:43 Shifting Focus from Accuracy to Business Value26:23 Leveraging PyTensor for Optimization34:27 Applying Optimal Decision-Making in Industry40:06 Understanding Utility Functions in Regulation41:35 Introduction to Obeisance Decision Theory Workflow42:33 Exploring Price Elasticity and Demand45:54 Optimizing Profit through Bayesian Models51:12 Risk Aversion and Utility Functions57:18 Advanced Risk Management Techniques01:01:08 Practical Applications of Bayesian Decision-Making01:06:54 Future Directions in Bayesian Inference01:10:16 The Quest for Better Inference Algorithms01:15:01 Dinner with a Polymath: Herbert SimonThank you to my Patrons for making this episode possible!Links from the show:Come meet Alex at the Field of Play Conference in Manchester, UK, March 27, 2026! https://www.fieldofplay.co.uk/A Bayesian decision theory workflowDaniel's website, LinkedIn and GitHubLBS #124 State Space Models & Structural Time Series, with Jesse GrabowskiLBS #123 BART & The Future of Bayesian Tools, with Osvaldo MartinLBS #74 Optimizing NUTS and Developing the ZeroSumNormal Distribution, with Adrian SeyboldtLBS #76 The Past, Present & Future of Stan, with Bob Carpenter

BITESIZE | How Do Diffusion Models Work?

Today's clip is from Episode 151 of the podcast, with Jonas ArrudaIn this conversation, Jonas Arruda explains how diffusion models generate data by learning to reverse a noise process. The idea is to start from a simple distribution like Gaussian noise and gradually remove noise until the target distribution emerges. This is done through a forward process that adds noise to clean parameters and a backward process that learns how to undo that corruption. A noise schedule controls how much noise is added or removed at each step, guiding the transformation from pure randomness back to meaningful structure.Get the full discussion here• Join this channel to get access to perks:https://www.patreon.com/c/learnbayesstats• Intro to Bayes Course (first 2 lessons free): https://topmate.io/alex_andorra/503302• Advanced Regression Course (first 2 lessons free): https://topmate.io/alex_andorra/1011122Our theme music is « Good Bayesian », by Baba Brinkman (feat MC Lars and Mega Ran). Check out his awesome work at https://bababrinkman.com/ !

151 Diffusion Models in Python, a Live Demo with Jonas Arruda

• Support & get perks!• Proudly sponsored by PyMC Labs! Get in touch at alex.andorra@pymc-labs.com• Intro to Bayes and Advanced Regression courses (first 2 lessons free)Our theme music is « Good Bayesian », by Baba Brinkman (feat MC Lars and Mega Ran). Check out his awesome work !Chapters:00:00 Exploring Generative AI and Scientific Modeling10:27 Understanding Simulation-Based Inference (SBI) and Its Applications15:59 Diffusion Models in Simulation-Based Inference19:22 Live Coding Session: Implementing Baseflow for SBI34:39 Analyzing Results and Diagnostics in Simulation-Based Inference46:18 Hierarchical Models and Amortized Bayesian Inference48:14 Understanding Simulation-Based Inference (SBI) and Its Importance49:14 Diving into Diffusion Models: Basics and Mechanisms50:38 Forward and Backward Processes in Diffusion Models53:03 Learning the Score: Training Diffusion Models54:57 Inference with Diffusion Models: The Reverse Process57:36 Exploring Variants: Flow Matching and Consistency Models01:01:43 Benchmarking Different Models for Simulation-Based Inference01:06:41 Hierarchical Models and Their Applications in Inference01:14:25 Intervening in the Inference Process: Adding Constraints01:25:35 Summary of Key Concepts and Future DirectionsThank you to my Patrons for making this episode possible!Links from the show:- Come meet Alex at the Field of Play Conference in Manchester, UK, March 27, 2026!- Jonas's Diffusion for SBI Tutorial & Review (Paper & Code)- The BayesFlow Library- Jonas on LinkedIn- Jonas on GitHub- Further reading for more mathematical details: Holderrieth & Erives- 150 Fast Bayesian Deep Learning, with David Rügamer, Emanuel Sommer & Jakob Robnik- 107 Amortized Bayesian Inference with Deep Neural Networks, with Marvin Schmitt

#150 Fast Bayesian Deep Learning, with David Rügamer, Emanuel Sommer & Jakob Robnik

• Support & get perks!• Proudly sponsored by PyMC Labs! Get in touch at alex.andorra@pymc-labs.com• Intro to Bayes and Advanced Regression courses (first 2 lessons free)Our theme music is « Good Bayesian », by Baba Brinkman (feat MC Lars and Mega Ran). Check out his awesome work !Chapters:00:00 Scaling Bayesian Neural Networks04:26 Origin Stories of the Researchers09:46 Research Themes in Bayesian Neural Networks12:05 Making Bayesian Neural Networks Fast16:19 Microcanonical Langevin Sampler Explained22:57 Bottlenecks in Scaling Bayesian Neural Networks29:09 Practical Tools for Bayesian Neural Networks36:48 Trade-offs in Computational Efficiency and Posterior Fidelity40:13 Exploring High Dimensional Gaussians43:03 Practical Applications of Bayesian Deep Ensembles45:20 Comparing Bayesian Neural Networks with Standard Approaches50:03 Identifying Real-World Applications for Bayesian Methods57:44 Future of Bayesian Deep Learning at Scale01:05:56 The Evolution of Bayesian Inference Packages01:10:39 Vision for the Future of Bayesian StatisticsThank you to my Patrons for making this episode possible!Come meet Alex at the Field of Play Conference in Manchester, UK, March 27, 2026!Links from the show:David Rügamer:* Website* Google Scholar* GitHubEmanuel Sommer:* Website* GitHub* Google ScholarJakob Robnik:* Google Scholar* GitHub* Microcanonical Langevin paper* LinkedIn