Activity for aadam.dev

Really enjoying the seamless self-hosted sync for the Logseq DB version (set up with the help of Codex). Logseq nowadays acts as both my daily journal, and a place for LLM to write logs and context to. I'm loving the experience so far. Thank you Logseq team for developing and open-sourcing sync.

Remember watching some demo for Discourse Graphs by @joelchan86.bsky.social years ago, tried it a bit in Logseq, but it didn't stick. Got back into it today after watching some talks on Open Modular Science. Still, a lot to explore, what's @atproto.science, @semble.so , and @leaflet.pub etc.

Such a timely find. I was just about to find some source to delve into the field HRL. I think this could be a good starting point.

Martin Klissarov@martinklissarov.bsky.social

As AI agents face increasingly long and complex tasks, decomposing them into subtasks becomes increasingly appealing. But how do we discover such temporal structure? Hierarchical RL provides a natural formalism-yet many questions remain open. Here's our overview of the field🧵

More and more exciting papers to sift through each day

Jean-Rémi King@jeanremiking.bsky.social

I'm very pleased to share our latest study: ‘Emergence of Language in the Developing Brain’, by L Evanson, P Bourdillon et al: - Paper: ai.meta.com/research/pub... - Blog: ai.meta.com/blog/meta-fa... - Thread below 👇

Excited to go through this paper

Hafez Ghaemi@hafezghm.bsky.social

Preprint Alert 🚀 Can we simultaneously learn transformation-invariant and transformation-equivariant representations with self-supervised learning? TL;DR Yes! This is possible via simple predictive learning & architectural inductive biases – without extra loss terms and predictors! 🧵 (1/10)