1 paper
Ji-Won Park, Chae Un Kim
In large-scale AI systems, allocating scarce resources such as GPU compute time and bandwidth among multiple agents is a critical challenge. Conventional policies focus on efficien…