Taija

Julia REPL which activates the Bluesky project, and calls a function which simulates creating a new post on the julialanguage Bluesky account!

Hello @bsky.app from the #JuliaLang community!

We're excited to start posting here on Bluesky in addition to our other social media, all of which can be found here: julialang.org

We'll keep you updated on our fantastic community, our conferences, #JuliaLang updates, and more on this profile!

New Package for Energy-Based Sampling – Taija A short post announcing a small new package: EnergySamplers.jl for energy-based sampling.

For an accessible introduction to the package, check out one of our recent blog posts: www.taija.org/blog/posts/n...

GitHub - JuliaTrustworthyAI/EnergySamplers.jl: A small Julia package for energy-based sampling. A small Julia package for energy-based sampling. Contribute to JuliaTrustworthyAI/EnergySamplers.jl development by creating an account on GitHub.

At the core, JEMs rely on energy-based posterior sampling, which we have implemented in github.com/JuliaTrustwo..., a light-weight package that extends Optimiser.jl.

Among other things, it powers JointEnergyModels.jl and helps us to estimate model posteriors for explainability purposes.

Did you know that classifiers have generative properties? Grathwohl et al. (2019) propose an approach for training joint energy-based models that learn to predict and generate: arxiv.org/abs/1912.03263. JEMs have great properties that lead to improved trustworthiness: arxiv.org/abs/2312.10648

Head over to our blog to learn more about some of the things Taija can do for you: www.taija.org/blog/

Hello Bluesky ☀️

Artificial Intelligence (AI) has been advancing rapidly in recent years. Consequently, Julia’s AI ecosystem has also been growing fast. Taija is an effort to provide users with tools to make sense of the AI models that they train and deploy.

Find out more here: www.taija.org