Learning Bayesian Statistics

What do neurodegenerative diseases, gerrymandering and ecological inference all have in common? Well, they can all be studied with Bayesian methods — and that’s exactly what Karin Knudson is doing.In this episode, Karin will share with us the vital and essential work she does to understand aspects of neurodegenerative diseases. She’ll also tell us more about computational neuroscience and Dirichlet processes — what they are, what they do, and when you should use them.Karin did her doctorate in mathematics, with a focus on compressive sensing and computational neuroscience at the University of Texas at Austin. Her doctoral work included applying hierarchical Dirichlet processes in the setting of neural data and focused on one-bit compressive sensing and spike-sorting.Formerly the chair of the math and computer science department of Phillips Academy Andover, she started a postdoc at Mass General Hospital and Harvard Medical in Fall 2019. Most importantly, rock climbing and hiking have no secrets for her!Our theme music is « Good Bayesian », by Baba Brinkman (feat MC Lars and Mega Ran). Check out his awesome work at https://bababrinkman.com/ ! Links from the show, personally curated by Karin Knudson:Karin on Twitter: https://twitter.com/karinknudsonSpike train entropy-rate estimation using hierarchical Dirichlet process priors (Knudson and Pillow): https://pillowlab.princeton.edu/pubs/abs_Knudson_HDPentropy_NIPS13.htmlFighting Gerrymandering with PyMC3, PyCon 2018, Colin Carroll and Karin Knudson: https://www.youtube.com/watch?v=G9I5ZnkWR0AExpository resources on Dirichlet Processes: Chapter 23 of Bayesian Data Analysis (Gelman et al.) and http://www.gatsby.ucl.ac.uk/~ywteh/research/npbayes/dp.pdfHierarchical Dirichlet Processes (introduced the HDP and included applications in topic modeling and for working with time-series data and Hidden Markov Models): https://www.stat.berkeley.edu/~aldous/206-Exch/Papers/hierarchical_dirichlet.pdfA Sticky HDP-HMM with applications to speaker diarization (a nice example of how the HDP can be used with HMM, in this case cleverly adapted so that states have more persistence): https://arxiv.org/abs/0905.2592If you want to get deeper into the weeds and also get a sense of the history: Dirichlet Processes with Applications to Bayesian Nonparametric Problems (https://projecteuclid.org/euclid.aos/1176342871) and A Bayesian Analysis of Some Nonparametric Problems (https://projecteuclid.org/euclid.aos/1176342360)

Colin Carroll discusses implementing Bayesian tools in finance and the airline industry, emphasizing effective communication to non-technical stakeholders. He explores challenges in model fitting with golfers' accuracy data, importance of pre-processing features, and practical applications of Bayesian methods. The future of probabilistic programming frameworks is also discussed, along with admiration for Professor Gilbert Strang.

Colin Carroll, a machine learning researcher and key contributor to PyMC3 and ArviZ, discusses the intricacies of probabilistic programming. He explains its value in the realm of Bayesian statistics and provides insights on selecting between various libraries like Stan and Pyro based on project requirements. Colin shares his journey from pure mathematics to data science and emphasizes the importance of quantifying uncertainty for better decision-making, particularly in high-stakes scenarios like flight insurance.

Chris Fonnesbeck, senior quantitative analyst for the New York Yankees and associate professor at Vanderbilt, dives deep into the world of Bayesian methods. He illustrates when to effectively employ these techniques and the challenges of teaching them. Fonnesbeck highlights their application in sports analytics, particularly in baseball, alongside marine biology findings. He discusses the importance of skills like programming and understanding priors, while also addressing issues like missing data, showcasing Bayesian's growing relevance across disciplines.

Meet Osvaldo Martin, a biologist, physicist, and data scientist discussing Bayesian methods and open-source development. Explore the intersection of bioinformatics and statistics, the influence of Bayesian methods, and the importance of model comparison in statistics and machine learning. Dive into ARVIS for visualizing Bayesian models and hypothetical dinners with scientific minds.

Are you a researcher or data scientist / analyst / ninja? Do you want to learn Bayesian inference, stay up to date or simply want to understand what Bayesian inference is? Well I'm just like you! When I started learning Bayesian methods, I really wished there were a podcast out there that could introduce me to the methods, the projects and the people who make all that possible.So I created "Learning Bayesian Statistics", a fortnightly podcast where I interview researchers and practitioners of all fields about why and how they use Bayesian statistics, and how in turn YOU, as a learner, can apply these methods in YOUR modeling workflow. Now the thing is, I’m not a beginner, but I’m not an expert either. The people I’ll interview will definitely be. So I’ll be learning alongside you. I won’t pretend to know everything in this podcast, and I WILL make mistakes. But thanks to the guests’ feedback, we’ll be able to learn from those mistakes, and I think this will help you (and me!) become better, faster, stronger Bayesians.So, whether you want to learn Bayesian statistics or hear about the latest libraries, books and applications, this podcast is for you. In this very first episode - well actually it’s episode 0, because 0-indexing rules! - I will introduce you to the genesis of this podcast, tell you why you should listen and reveal some of the guests for the coming episodes.Come join us!Links from the show:Podcast website: https://learnbayesstats.anvil.app/Alex Twitter feed: https://twitter.com/alex_andorra