A Practical Guide to Autonomous Evaluation Loops in Claude Code

The guide explains two layers of Claude Code improvement, YAML activation tuning and output checks like word count and sentence rules.
IEEE

Feasible Policy Iteration With Guaranteed Safe Exploration

Abstract: Safety guarantee is an important topic when training real-world tasks with reinforcement learning (RL). During online environmental exploration, any constraint violation can lead to ...
IEEE

Data-Driven Reinforcement Learning-Based Forgetting Factor Iterative Learning Control

Abstract: In this paper, a data-driven control scheme is proposed by integrating both reinforcement learning (RL) and iterative learning control (ILC) methodologies. To address the limitations of ...