Lukas Schumacher's Avatar

Lukas Schumacher

@schumacherlu.bsky.social

Postdoctoral Researcher at University of Basel Interested in Cognitive Modeling | Decision Making | Dynamics in Cognition | Amortized Bayesian Inference | Superstatistics

196 Followers  |  377 Following  |  2 Posts  |  Joined: 04.11.2024  |  1.5547

Latest posts by schumacherlu.bsky.social on Bluesky

Introduction โ€“ Amortized Bayesian Cognitive Modeling

๐Ÿง  Check out the classic examples from Bayesian Cognitive Modeling: A Practical Course (Lee & Wagenmakers, 2013), translated into step-by-step tutorials with BayesFlow!

Interactive version: kucharssim.github.io/bayesflow-co...

PDF: osf.io/preprints/ps...

30.05.2025 14:28 โ€” ๐Ÿ‘ 29    ๐Ÿ” 14    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
Post image

A study with 5M+ data points explores the link between cognitive parameters and socioeconomic outcomes: The stability of processing speed was the strongest predictor.

BayesFlow facilitated efficient inference for complex decision-making models, scaling Bayesian workflows to big data.

๐Ÿ”—Paper

03.02.2025 12:21 โ€” ๐Ÿ‘ 19    ๐Ÿ” 6    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

Paul Bรผrkner (TU Dortmund University), will give our next talk. This will be about "Amortized Mixture and Multilevel Models", and is scheduled on Thursday the 30th January at 11am. To receive the link to join, sign up at listserv.csv.warwick...

14.01.2025 12:00 โ€” ๐Ÿ‘ 16    ๐Ÿ” 4    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 3

making your own sourdough starter is pretty simple. Give it a go :)

10.01.2025 07:07 โ€” ๐Ÿ‘ 2    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

Looks amazing. I recently started to bake with sourdough but with much less success...

09.01.2025 15:18 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

Check out this project on modeling stationary and time-varying parameters with BayesFlow.

The family of methods is called "neural superstatistics", how can it not be cool!? ๐Ÿ˜Ž

๐Ÿ‘จโ€๐Ÿ’ป Led by @schumacherlu.bsky.social

06.12.2024 12:25 โ€” ๐Ÿ‘ 11    ๐Ÿ” 3    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
Post image

BayesFlow is a library for amortized Bayesian inference with neural networks.

โ‹… Multi-backend via Keras 3: Use PyTorch, TensorFlow, or JAX.
โ‹… Modern nets: Flow matching, diffusion, consistency models, normalizing flows, transformers
โ‹… Built-in diagnostics and plotting

๐Ÿ”— github.com/bayesflow-or...

22.11.2024 22:30 โ€” ๐Ÿ‘ 107    ๐Ÿ” 22    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

@schumacherlu is following 20 prominent accounts