rl-wrong-about-rewards.md

kept by eddie dec 08, 25

The standard reinforcement learning formalization errs by placing reward in the environment; instead, reward should be part of the agent, enabling goal-driven behavior.

Claude Agent Skills: A First Principles Deep Dive
Claude's Agent Skills use a prompt-based meta-tool architecture that injects specialized instructions and context modifications to guide complex workflows without executing code.
also on #agent-design
On Building Agents From First Principles (via @athleticKoder)
A step-by-step methodology for building AI agents by defining environments, generating teacher trajectories, fine-tuning student models, and iteratively improving them with reinforcement learning.
also on #reinforcement-learning, #machine-learning, #ai-research
[2604.27505] Leveraging Verifier-Based Reinforcement Learning in Image Editing
Edit-R1 uses a chain-of-thought reasoning reward model, trained with Group Contrastive Preference Optimization, to provide fine-grained, interpretable rewards that improve image-editing model performance.
also on #reinforcement-learning, #machine-learning, #ai-research