Tagged reinforcement learning · show all articles
Grading a model once meant scoring a string against a reference answer. Grading an agent required borrowing a games lab's tools, discarding them, and rebuilding evaluation from scratch, in a documented, dated sequence of its own.
Five structural problems in agent design remain open by the field's own account — long-horizon credit assignment, durable state, multi-agent coordination cost, action verification, and silent sub-task failure.
Every era of AI agent architecture answers one question differently: what decides the next action. A 1971 planner, a learned reward signal, a prompted reasoning loop, a delegated sub-agent, and a shipped SDK are five dated, sourced answers.
The line from Shannon's letter counts to trained deliberation is usually told as one long scaling curve. It was not. It was a sequence of specific bottlenecks, each named, each broken by a different idea.