Speaker
Tengpeng Xu
(Sun Yat-sen University)
Description
Driven by the need for fast and efficient algorithms in cosmic statistics to handle the massive data volumes from ongoing and forthcoming galaxy surveys, we present the first public release of Hermes (HypER-speed MultiREsolution cosmic Statistics) — an open-source, parallelized, and GPU-accelerated Python package. At its core, Hermes implements the MRA-CS (MultiResolution Analysis for Cosmic Statistics) framework. The talk will provide a concise overview of the mathematical foundations and core algorithms behind Hermes, demonstrate its performance through numerical tests, and highlight its broad potential for extracting cosmological information from next-generation galaxy surveys.
Authors
Prof.
Longlong Feng
(Sun Yat-sen University)
Tengpeng Xu
(Sun Yat-sen University)