$\alpha^{\alpha}$-Rank: Practically Scaling $\alpha$-Rank through Stochastic Optimisation

TL;DR

This paper introduces $ ext{ extalpha}^{ ext{ extalpha}}$-Rank, a scalable stochastic optimization method that significantly improves the practicality of ranking strategies in large multi-agent systems, overcoming previous computational limitations.

Contribution

The authors develop $ ext{ extalpha}^{ ext{ extalpha}}$-Rank, a stochastic implementation that reduces complexity and enables large-scale multi-agent evaluation, addressing the infeasibility of original $ ext{ extalpha}$-Rank.

Findings

Achieves 1000x speedup on large random matrices

Enables evaluation of joint strategies with up to 33 million profiles

Demonstrates practical scalability in large multi-agent settings

Abstract

Recently, $\alpha$-Rank, a graph-based algorithm, has been proposed as a solution to ranking joint policy profiles in large scale multi-agent systems. $\alpha$-Rank claimed tractability through a polynomial time implementation with respect to the total number of pure strategy profiles. Here, we note that inputs to the algorithm were not clearly specified in the original presentation; as such, we deem complexity claims as not grounded, and conjecture solving $\alpha$-Rank is NP-hard. The authors of $\alpha$-Rank suggested that the input to $\alpha$-Rank can be an exponentially-sized payoff matrix; a claim promised to be clarified in subsequent manuscripts. Even though $\alpha$-Rank exhibits a polynomial-time solution with respect to such an input, we further reflect additional critical problems. We demonstrate that due to the need of constructing an exponentially large Markov chain, $\alpha$-Rank is infeasible beyond a small finite number of agents. We ground these claims by adopting amount of dollars spent as a non-refutable evaluation metric. Realising such scalability issue, we present a stochastic implementation of $\alpha$-Rank with a double oracle mechanism allowing for reductions in joint strategy spaces. Our method, $\alpha^\alpha$-Rank, does not need to save exponentially-large transition matrix, and can terminate early under required precision. Although theoretically our method exhibits similar worst-case complexity guarantees compared to $\alpha$-Rank, it allows us, for the first time, to practically conduct large-scale multi-agent evaluations. On $10^4 \times 10^4$ random matrices, we achieve $1000x$ speed reduction. Furthermore, we also show successful results on large joint strategy profiles with a maximum size in the order of $\mathcal{O}(2^{25})$ ($\approx 33$ million joint strategies) -- a setting not evaluable using $\alpha$-Rank with reasonable computational budget.