Scalable Option Learning Accelerates Hierarchical Reinforcement Learning

Scalable Option Learning (SOL) algorithm boosts hierarchical RL training throughput by about 35× and was tested on 30 billion frames in NetHack, outperforming flat agents. getnews.me/scalable-option-learning... #hierarchicalrl #sol #nethack

Hierarchical RL Boosts Intraday Risk Factor Generation

A hierarchical RL system (HPPO) auto-creates intraday risk factors, delivering a 25% excess return in back-tests on the CSI 300, CSI 800, Nifty 100 and S&P 500. Read more: getnews.me/hierarchical-rl-boosts-i... #hierarchicalrl #intraday

Hierarchical RL Enables Collaborative Pick‑and‑Place with Legged Robots

A hierarchical RL pipeline boosts training efficiency ~55% and cuts execution time 18.6% for one‑armed legged robots doing pick‑and‑place, proven on ANYmal D in dual trials. Read more: getnews.me/hierarchical-rl-enables-... #hierarchicalrl #anymal