Max Hinne

Just published in JOSS: 'Bamojax: Bayesian Modelling with JAX' https://doi.org/10.21105/joss.08642

It's actually been super useful for us in several GP related models!

GitHub - UncertaintyInComplexSystems/bamojax: Bayesian Modelling using Jax Bayesian Modelling using Jax. Contribute to UncertaintyInComplexSystems/bamojax development by creating an account on GitHub.

Bamojax (Bayesian modelling with JAX) v0.3.10 is out! Next to increased support for distributions and bijectors, several marginal likelihood estimators are now available, such as bridge sampling and truncated harmonic mean estimation (THAMES).

Check it out on github.com/UncertaintyI...!

Very proud on David Leeftink's first PhD thesis paper on optimal control of probabilistic dynamics models! Check it out here: arxiv.org/abs/2504.02543

After five years of confused staring at Greek letters, it is my absolute pleasure to finally share our (with @smfleming.bsky.social) computational model of mental imagery and reality monitoring: Perceptual Reality Monitoring as Higher-Order inference on Sensory Precision ✨
osf.io/preprints/ps...

Teken de petitie: Stop de asielwetten | GroenLinks-PvdA Stop de asielwetten! Deze wetten van het kabinet Wilders-Yesilgöz zijn asociaal, onverantwoord en onmenselijk. Teken de petitie en hou ze tegen.

Na een lange en chaotische avond, stemde de Tweede Kamer voor de asielwetten van minister Faber.

Het wordt zélfs strafbaar om ongedocumenteerde mensen te helpen met onderdak, een maaltijd of kleding.

Help ons dit beleid tegen te houden, teken én deel de petitie:

groenlinkspvda.nl/petitie/stop...

I am in downtown Los Angeles right now and it is completely surreal to hear that Trump is sending Marines here. We are listening to mariachi music. People have dogs. There are teenagers here. It is utterly and completely peaceful. Please tell your friends.

Job Alert! Two PhD positions in the Probabilistic Graphical Models group of the @DondersInst at @RadboudU. Both PhDs are part of the ELSA lab on Legal, Regulatory, and Policy Aspects of Clinical Decision Support Systems. 1/8

Bayes-by's first Markov chain? :o)

Picture of the abstract of our new paper

Picture of our key results (Figure 1)

🎉 Effects of climate protests by scientists paper published!

Our findings suggest that scientists can engage in public protest without compromising their credibility, but that such actions alone may have less impact than one would like to believe: royalsocietypublishing.org/doi/10.1098/...

My lab is looking for a PhD candidate, in collaboration with ​Bernhard Englitz​'s lab.

www.ru.nl/en/working-a...

New report reveals key role universities play in boosting growth and productivity across the UK The higher education sector contributes more than £265 billion to the UK economy.

"...for every £1 of public money invested into UK universities, £14 of economic benefit is generated" www.universitiesuk.ac.uk/latest/news/...

Long COVID activists fought Trump team’s research cuts and won ― for now Success gives hope to scientists and advocates who managed to get millions of dollars in grants restored.

Long COVID advocates and researchers in the United States have done the extraordinary. After a bruising battle, they managed to revive some of the research grants cancelled by the administration of President Donald Trump.

https://go.nature.com/3RKg5tP

David Leeftink's first PhD thesis work is now available on arXiv! Probabilistic Pontryagin's Maximum Principle for Continuous-Time Model-Based Reinforcement Learning, check it out!

Model-Based Neuroscience and Cognition Summer School Visit the post for more.

Join us in Amsterdam for the 11th Model-Based Neuroscience & Cognition Summer School. Gain a solid foundation in Bayesian evidence-accumulation modeling and explore its integration with neuroscience or advanced computational frameworks, including reinforcement-learning models. modelbasedneurosci.com

An introduction to Sequential Monte Carlo for Bayesian inference and model comparison—with examples for psychology and behavioral science - Behavior Research Methods Bayesian inference is becoming an increasingly popular framework for statistics in the behavioral sciences. However, its application is hampered by its computational intractability – almost all Bayesi...

My tutorial on Sequentual Monte Carlo for psychology and behavioural science is out now in Behavior Research Methods! Check it out at link.springer.com/article/10.3...

7/ Much more interesting examples, ranging from variable selection to Wishart processes, can be found at https://github.com/UncertaintyInComplexSystems/bamojax/tree/main/bamojax/examples

Happy modelling!

6/ model = Model('Gaussian with unknown mean')

unknown_mean = model.add_node('mu', distribution=dx.Normal(loc=mu0, scale=sd0))

y = model.add_node('y', distribution=dx.Normal, observations=y, parents=dict(loc=unknown_mean, scale=true_sd))

5/ Here is a very simple model, where we estimate the mean of a Gaussian using bamojax. We use Distrax to define probability distributions:

4/ In contrast to deriving a log-density from e.g. PyMC and using Blackjax to sample from it, with *bamojax* you have fine-grained control over individual Gibbs steps, which leads to much more efficient inference.

3/ This makes it possible to have *both* fast modelling, and fast inference!

2/ Very fast Bayesian inference in JAX is possible through Blackjax, but model development is easier in probabilistic programming languages like Stan and PyMC. *bamojax* provides automated Gibbs sampling, using the different sampling algorithms provided by Blackjax to update individual variables.

🧵 Introducing *bamojax* v0.1 --- Bayesian Modelling in JAX! https://github.com/uncertaintyincomplexsystems/bamojax

Ufff this went from bad to downright horrifying in less than a week :'-(