2020/01/14 by David Balduzzi, Wojciech M Czarnecki, Thomas W Anthony +5 · 1 voice
Computer Science · Mathematics · #cs.AI #cs.GT #cs.LG #cs.MA #stat.ML
published as ICLR 2020 · 18 pages, 3 figures
arxiv created 2020/01/18 · arxiv updated 2020/01/22
With the success of modern machine learning, it is becoming increasingly important to understand and control how learning algorithms interact. Unfortunately, negative results from game theory show there is little hope of understanding or controlling general n-player games. We therefore introduce smooth markets (SM-games), a class of n-player games with pairwise zero sum interactions. SM-games codify a common design pattern in machine learning that includes (some) GANs, adversarial training, and other recent algorithms. We show that SM-games are amenable to analysis and optimization using first-order methods.