Microsoft Artificial Intelligence
Microsoft Artificial Intelligence
4 years ago

Combining No-regret and Q-learning

Counterfactual Regret Minimization (CFR) has found success in settings like poker which have both terminal states and perfect recall. We seek to understand how to relax these requirements. As a first step, we introduce a simple algorithm, local no-regret learning (LONR), which uses a Q-learning-like update rule to allow learning without terminal states or perfect recall. We prove its convergence for the basic case of MDPs (where Q-learning already suffices), as well as limited extensions of them. With a straightforward modification, we extend the basic premise of LONR to work in multi-agent settings and present empirical results showing that it achieves last iterate convergence in a number of settings. Most notably, we show this for NoSDE games, a class of Markov games specifically designed to be impossible for Q-value-based methods to learn and where no prior algorithm is known to achieve convergence to a stationary equilibrium even on average. Furthermore, by leveraging last iterate converging no-regret algorithms (one of which we introduce), we show empirical last iterate convergence in all domains tested with LONR.

Publisher URL: https://www.microsoft.com/en-us/research/publication/combining-no-regret-and-q-learning-2/

DOI: microsoft_corporation.25090.129

You might also like
Discover & Discuss Important Research

Keeping up-to-date with research can feel impossible, with papers being published faster than you'll ever be able to read them. That's where Researcher comes in: we're simplifying discovery and making important discussions happen. With over 19,000 sources, including peer-reviewed journals, preprints, blogs, universities, podcasts and Live events across 10 research areas, you'll never miss what's important to you. It's like social media, but better. Oh, and we should mention - it's free.

  • Download from Google Play
  • Download from App Store
  • Download from AppInChina

Researcher displays publicly available abstracts and doesn’t host any full article content. If the content is open access, we will direct clicks from the abstracts to the publisher website and display the PDF copy on our platform. Clicks to view the full text will be directed to the publisher website, where only users with subscriptions or access through their institution are able to view the full article.