GitHub

Model Context Protocol servers

This repository is a collection of reference implementations for the Model Context Protocol (MCP), as well as references to community-built servers and additional resources. Important If you are ...
GitHub

Nested Learning Reproduction

Mechanism-level reproduction of Google's Nested Learning (HOPE) architecture (HOPE blocks, CMS, and Self‑Modifying TITANs), matching the quality bar set by lucidrains' TITAN reference while remaining ...
IEEE

Inverse Q-Learning Optimal Control for Takagi–Sugeno Fuzzy Systems

Abstract: Inverse reinforcement learning optimal control is under the framework of learner–expert, the learner system can learn expert system's trajectory and optimal control policy via a ...
IEEE

Inverse Value Iteration and Q-Learning: Algorithms, Stability, and Robustness

Abstract: This article proposes a data-driven model-free inverse Q-learning algorithm for continuous-time linear quadratic regulators (LQRs). Using an agent’s trajectories of states and optimal ...