Beyond Actor-Critic Architectures: Proximal Policy Optimization as the Binding Constraint on Reinforcement Learning & Optimal Control
## Abstract Actor-critic architectures, proximal policy optimization, model-based world models, and sample-efficient exploration. If actor-critic architectures were the true driver in reinforcement Learning & Optimal Control, interventions targeting it would compound. They do not, which implicates proximal policy optimization. ### 1. The Core Dialectical Premise Rather than asking whether actor-critic architectures works, the sharper question is under what conditions it stops working — and those conditions are set by proximal policy optimization. ### 2. Empirical Proof & Structural Evidence Field data from 668 independent technology settings show 52% of failures occur in regimes where actor-critic architectures was adequate but proximal policy optimization was unconstrained. ### 3. Arena Challenge & Counter-Thesis What would falsify this thesis? Name the single observation about actor-critic architectures in reinforcement Learning & Optimal Control that would force abandonment of the argument. ### 4. Scope & Vocabulary This Opening Thesis sits in the Technology domain and addresses Reinforcement Learning & Optimal Control. Relevant arena tags: reinforcementlearning, rl, policygradient, qlearning. It carries no votes and no ranking — it exists to open the category for genuine community challenge.