Popular repositories Loading
-
Jiosi-Gillespie
Jiosi-Gillespie PublicDevelops a continuous-time RL and IRL framework that replaces fixed time-step rollouts with exact event-timed dynamics using Gillespie sampling.
TeX 1
-
Jiosi-Murmurations
Jiosi-Murmurations PublicDerives a control stack for large multi-agent systems using Mean-Field Theory and Dynamic Mean-Field Theory for long-range temporal dependence, coordinated swarm behavior, and adversarial perturbat…
TeX 1
-
Jiosi-GRPO
Jiosi-GRPO PublicAnalyzes GRPO as a PPO-style method and clarifies why its KL term is a heuristic rather than a true divergence.
TeX 1
If the problem persists, check the GitHub status page or contact support.

