))
Effort.jl: Fast, Differentiable Cosmology on a Laptop
Manage episode 517698281 series 3690682
Content provided by Mike Breault. All podcast content including episodes, graphics, and podcast descriptions are uploaded and provided directly by Mike Breault or their podcast platform partner. If you believe someone is using your copyrighted work without your permission, you can follow the process outlined here https://podcastplayer.com/legal.
We explore how effort.jl turns petabytes of cosmology data into fast, trustworthy inferences. A fast neural-network surrogate and physics-informed preprocessing deliver ~15 microseconds per spectrum on a single CPU, enabling gradient-based samplers like HMC/NUTS via Turing.jl to converge in minutes on a laptop. Validated against PT Challenge and BOSS data, the approach preserves accuracy and opens doors for cross-disciplinary applications in weather, climate, and materials.
…
continue reading
Note: This podcast was AI-generated, and sometimes AI can make mistakes. Please double-check any critical information.
Sponsored by Embersilk LLC
1435 episodes
Share
