Vincent Fortuin, a tenure-track research leader at Helmholtz AI, dives into the world of Bayesian deep learning and its transformative power in scientific research. He discusses the contradictions in AI hype versus practical outcomes and highlights how combining Bayesian statistics with deep learning enhances prediction reliability. Fortuin emphasizes the significance of prior knowledge and argues for better uncertainty communication in AI applications. He also touches on innovative techniques that improve data efficiency and acknowledges the need for collaboration in the community.

Matt Penn, a football data scientist for Como 1907, discusses the critical role of Bayesian statistics in player recruitment and team decisions. He highlights how analysts can effectively communicate insights to coaching staff, fostering better collaboration. The conversation dives into the biases present in traditional football statistics and the impact of tracking data on understanding player movements. Matt shares his journey into sports analytics and emphasizes the importance of curiosity and practical experience for aspiring analysts in this competitive landscape.

Proudly sponsored by PyMC Labs, the Bayesian Consultancy. Book a call, or get in touch!Intro to Bayes Course (first 2 lessons free)Advanced Regression Course (first 2 lessons free)Our theme music is « Good Bayesian », by Baba Brinkman (feat MC Lars and Mega Ran). Check out his awesome work!Visit our Patreon page to unlock exclusive Bayesian swag ;)Thank you to my Patrons for making this episode possible!Yusuke Saito, Avi Bryant, Ero Carrera, Giuliano Cruz, Tim Gasser, James Wade, Tradd Salvo, William Benton, James Ahloy, Robin Taylor,, Chad Scherrer, Zwelithini Tunyiswa, Bertrand Wilden, James Thompson, Stephen Oates, Gian Luca Di Tanna, Jack Wells, Matthew Maldonado, Ian Costley, Ally Salim, Larry Gill, Ian Moran, Paul Oreto, Colin Caprani, Colin Carroll, Nathaniel Burbank, Michael Osthege, Rémi Louf, Clive Edelsten, Henri Wallen, Hugo Botha, Vinh Nguyen, Marcin Elantkowski, Adam C. Smith, Will Kurt, Andrew Moskowitz, Hector Munoz, Marco Gorelli, Simon Kessell, Bradley Rode, Patrick Kelley, Rick Anderson, Casper de Bruin, Philippe Labonde, Michael Hankin, Cameron Smith, Tomáš Frýda, Ryan Wesslen, Andreas Netti, Riley King, Yoshiyuki Hamajima, Sven De Maeyer, Michael DeCrescenzo, Fergal M, Mason Yahr, Naoya Kanai, Steven Rowland, Aubrey Clayton, Jeannine Sue, Omri Har Shemesh, Scott Anthony Robson, Robert Yolken, Or Duek, Pavel Dusek, Paul Cox, Andreas Kröpelin, Raphaël R, Nicolas Rode, Gabriel Stechschulte, Arkady, Kurt TeKolste, Gergely Juhasz, Marcus Nölke, Maggi Mackintosh, Grant Pezzolesi, Avram Aelony, Joshua Meehl, Javier Sabio, Kristian Higgins, Alex Jones, Gregorio Aguilar, Matt Rosinski, Bart Trudeau, Luis Fonseca, Dante Gates, Matt Niccolls, Maksim Kuznecov, Michael Thomas, Luke Gorrie, Cory Kiser, Julio, Edvin Saveljev, Frederick Ayala, Jeffrey Powell, Gal Kampel, Adan Romero, Will Geary, Blake Walters, Jonathan Morgan, Francesco Madrisotti, Ivy Huang, Gary Clarke, Robert Flannery, Rasmus Hindström, Stefan, Corey Abshire, Mike Loncaric, David McCormick, Ronald Legere, Sergio Dolia and Michael Cao.Takeaways:Sharks play a crucial role in maintaining healthy ocean ecosystems.Bayesian statistics are particularly useful in data-poor environments like ecology.Teaching Bayesian statistics requires a shift in mindset from traditional statistical methods.The shark meat trade is significant and often overlooked.Ray meat trade is as large as shark meat trade, with specific markets dominating.Understanding the ecological roles of species is essential for effective conservation.Causal language is important in ecological research and should be encouraged.Evidence-driven decision-making is crucial in balancing human and ecological needs.Expert opinions are crucial for understanding species composition in landings.Trade dynamics are influenced by import preferences and species availability.Bayesian modeling allows for the incorporation of various data sources and expert knowledge.Field data collection is essential for validating model assumptions.The complexity of trade relationships necessitates a nuanced approach to modeling.Understanding the impact of management interventions on landings is critical.The role of scientists in informing policy is vital for effective conservation efforts.Chapters:00:00 Introduction to Marine Biology and Statistics04:33 The Role of Bayesian Statistics in Marine Research10:09 Challenges in Teaching Bayesian Statistics21:58 The Importance of Sharks in Ecosystems26:35 Understanding Shark Meat Trade and Conservation32:09 The Trade in Ray and Shark Meat36:18 Modeling Landings and Trade42:56 Challenges in Data Integration44:50 Running Complex Models51:57 Expert Elicitation and Prior Construction55:52 Future Directions and Research56:46 Reflections on Science and PolicyLinks from the show:Fisheries Lab: https://ifisheries.org/?page_id=83LBS #51 Bernoulli’s Fallacy & the Crisis of Modern Science, with Aubrey Clayton: https://learnbayesstats.com/episode/51-bernoullis-fallacy-crisis-modern-science-aubrey-claytonTranscriptThis is an automatic transcript and may therefore contain errors. Please get in touch if you're willing to correct them.

Join Will Dean, a statistician and data scientist at PyMC Labs, as he dives into the world of Bayesian marketing analytics. He shares insights on leveraging customer lifetime value (CLV) and media mix modeling to optimize marketing strategies. The discussion highlights the significance of productionizing models and the challenges that come with it, including version control and model management. With a focus on open-source collaboration, Will emphasizes the importance of continuous learning and innovative approaches to empower marketers with actionable data.

Chris Fonnesbeck, a trailblazer in sports analytics and a core developer of PyMC, shares insights from his unique journey that blends marine biology and statistical consulting with sports modeling. He discusses the evolution of sports analytics, particularly in baseball, and highlights the impact of Bayesian methods. Fonnesbeck dives into advanced modeling techniques using Gaussian processes and emphasizes community contributions to the future of PyMC. With humor, he relates the complexities of polling data to sports analytics, showing the fun side of data interpretation.

Jesse Grabowski, PhD candidate at Paris 1 Pantheon-Sorbonne and principal data scientist at PyMC Labs, dives into the intricate world of state space models in time series analysis. He discusses the powerful adaptability of Bayesian methods in econometrics, emphasizing how they enhance forecasting accuracy. Grabowski highlights the balance between model complexity and simplicity, the significance of understanding trends, and the practical applications of innovations and latent states. Plus, he unwraps the role of the Kalman filter in managing dynamic data.

Osvaldo Martin, a collaborator on various open-source Bayesian projects and educator at PyMC Labs, discusses the power of Bayesian Additive Regression Trees (BART). He explains how BART simplifies modeling for those lacking domain expertise. The conversation also highlights advancements in tools like PyMC-BART and PreliZ, emphasizing their contributions to prior elicitation. Osvaldo shares insights on integrating BART with Bambi and the importance of interactive learning in teaching Bayesian statistics. Additionally, he touches upon future enhancements for user experience in Bayesian analysis.

Hugo Bowne-Anderson is an independent data and AI consultant, known for his insights in education and podcasting. He discusses the challenges of building and deploying Large Language Model applications, stressing the importance of feedback in data science education. Hugo shares his journey in creating practical courses, highlighting the need for continuous learning in tech. He emphasizes collaboration between data scientists and software engineers and reflects on the evolving landscape of AI, calling for problem-solving over specific tools for aspiring data scientists.

Nathaniel Forde, a staff data scientist at Personio and a contributor to the PyMC ecosystem, shares insights on Bayesian structural equation modeling. He highlights the importance of confirmatory factor analysis in validating constructs and discusses the flexibility of Bayesian methods in analyzing complex relationships. Forde emphasizes the necessity of model validation and sensitivity analysis to ensure robust findings. He also reflects on his journey into data science, noting how early challenges shaped his approach to data quality and causal inference.

Proudly sponsored by PyMC Labs, the Bayesian Consultancy. Book a call, or get in touch!My Intuitive Bayes Online Courses1:1 Mentorship with meOur theme music is « Good Bayesian », by Baba Brinkman (feat MC Lars and Mega Ran). Check out his awesome work!Visit our Patreon page to unlock exclusive Bayesian swag ;)-------------------------Love the insights from this episode? Make sure you never miss a beat with Chatpods! Whether you're commuting, working out, or just on the go, Chatpods lets you capture and summarize key takeaways effortlessly.Save time, stay organized, and keep your thoughts at your fingertips.Download Chatpods directly from App Store or Google Play and use it to listen to this podcast today!https://www.chatpods.com/?fr=LearningBayesianStatistics-------------------------Takeaways:Epidemiology focuses on health at various scales, while biology often looks at micro-level details.Bayesian statistics helps connect models to data and quantify uncertainty.Recent advancements in data collection have improved the quality of epidemiological research.Collaboration between domain experts and statisticians is essential for effective research.The COVID-19 pandemic has led to increased data availability and international cooperation.Modeling infectious diseases requires understanding complex dynamics and statistical methods.Challenges in coding and communication between disciplines can hinder progress.Innovations in machine learning and neural networks are shaping the future of epidemiology.The importance of understanding the context and limitations of data in research. Chapters:00:00 Introduction to Bayesian Statistics and Epidemiology03:35 Guest Backgrounds and Their Journey10:04 Understanding Computational Biology vs. Epidemiology16:11 The Role of Bayesian Statistics in Epidemiology21:40 Recent Projects and Applications in Epidemiology31:30 Sampling Challenges in Health Surveys34:22 Model Development and Computational Challenges36:43 Navigating Different Jargons in Survey Design39:35 Post-COVID Trends in Epidemiology42:49 Funding and Data Availability in Epidemiology45:05 Collaboration Across Disciplines48:21 Using Neural Networks in Bayesian Modeling51:42 Model Diagnostics in Epidemiology55:38 Parameter Estimation in Compartmental ModelsThank you to my Patrons for making this episode possible!Yusuke Saito, Avi Bryant, Ero Carrera, Giuliano Cruz, Tim Gasser, James Wade, Tradd Salvo, William Benton, James Ahloy, Robin Taylor,, Chad Scherrer, Zwelithini Tunyiswa, Bertrand Wilden, James Thompson, Stephen Oates, Gian Luca Di Tanna, Jack Wells, Matthew Maldonado, Ian Costley, Ally Salim, Larry Gill, Ian Moran, Paul Oreto, Colin Caprani, Colin Carroll, Nathaniel Burbank, Michael Osthege, Rémi Louf, Clive Edelsten, Henri Wallen, Hugo Botha, Vinh Nguyen, Marcin Elantkowski, Adam C. Smith, Will Kurt, Andrew Moskowitz, Hector Munoz, Marco Gorelli, Simon Kessell, Bradley Rode, Patrick Kelley, Rick Anderson, Casper de Bruin, Philippe Labonde, Michael Hankin, Cameron Smith, Tomáš Frýda, Ryan Wesslen, Andreas Netti, Riley King, Yoshiyuki Hamajima, Sven De Maeyer, Michael DeCrescenzo, Fergal M, Mason Yahr, Naoya Kanai, Steven Rowland, Aubrey Clayton, Jeannine Sue, Omri Har Shemesh, Scott Anthony Robson, Robert Yolken, Or Duek, Pavel Dusek, Paul Cox, Andreas Kröpelin, Raphaël R, Nicolas Rode, Gabriel Stechschulte, Arkady, Kurt TeKolste, Gergely Juhasz, Marcus Nölke, Maggi Mackintosh, Grant Pezzolesi, Avram Aelony, Joshua Meehl, Javier Sabio, Kristian Higgins, Alex Jones, Gregorio Aguilar, Matt Rosinski, Bart Trudeau, Luis Fonseca, Dante Gates, Matt Niccolls, Maksim Kuznecov, Michael Thomas, Luke Gorrie, Cory Kiser, Julio, Edvin Saveljev, Frederick Ayala, Jeffrey Powell, Gal Kampel, Adan Romero, Will Geary, Blake Walters, Jonathan Morgan, Francesco Madrisotti, Ivy Huang, Gary Clarke, Robert Flannery, Rasmus Hindström and Stefan.Links from the show:LBS #21, Gaussian Processes, Bayesian Neural Nets & SIR Models, with Elizaveta Semenova: https://learnbayesstats.com/episode/21-gaussian-processes-bayesian-neural-nets-sir-models-with-elizaveta-semenova/Liza’s website: https://www.elizaveta-semenova.com/Liza on GitHub: https://github.com/elizavetasemenovaLiza on LinkedIn: https://www.linkedin.com/in/elizaveta-semenova/Liza on Google Scholar: https://scholar.google.com/citations?user=jqGIgFEAAAAJ&hl=enChris' page: https://www.bdi.ox.ac.uk/Team/c-wymantChris on GitHub: https://github.com/chrishivChris on LinkedIn: https://www.linkedin.com/in/chris-wymant-65661274/Chris on Blue Sky: https://bsky.app/profile/chriswymant.bsky.socialChris on Google Scholar: https://scholar.google.com/citations?user=OJ6t2UwAAAAJ&hl=enPriorVAE Paper: Explains how to build an emulator for a GP using a deep generative model (Variational Autoencoder, or VAE) and apply it within MCMC. Link to the paperPriorCVAE Paper: Builds on PriorVAE by encoding model parameters along with emulating stochastic process realisations. Includes examples for GPs, ODEs, and double-well models. Link to the paperStanCon 2024 Tutorial: A tutorial covering the basics of sequential decision-making, with a demo of Bayesian Optimization using Stan. Link to the tutorialNumpyro Course: Materials from a course Liza taught -- great for learning Numpyro. Link to the courseaggVAE Paper: An application of PriorVAE to the problem of changing boundaries. Link to the paperTranscriptThis is an automatic transcript and may therefore contain errors. Please get in touch if you're willing to correct them.