On the Rate of Convergence of Fictitious Play
By Felix Brandt, Felix A. Fischer, Paul Harrenstein. 2010
36 citations recorded by the source corpus.
Subfields: Management Science and Operations Research, Safety Research
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Management Science and Operations Research
- Alekh Agarwal
- J. Langford
- Chen-Yu Wei
- Marc-André Zöller
- Marco F. Huber
- Qinxuan Pan
- Constantinos Daskalakis
- Nathan Kallus
In Safety Research
- Marc-André Zöller
- Marco F. Huber
- Qinxuan Pan
- Constantinos Daskalakis
- Alekh Agarwal
- Dana Pessach
- Erez Shmueli
- Harikrishna Narasimhan
Citations in the corpus
- Survey on Automated Machine Learning (2019, methodology): “A common drawback of self-play approaches is the low convergence speed (Brandt et al., 2010) making this approach rather unsuited for AutoML.”
- A Counter-example to Karlin's Strong Conjecture for Fictitious Play (2014, methodology): “So, after exchanging the strategies 1 ↔ 2, we can repeat the players’ choices at Steps 3, 4, 5 and 6 in Steps 7, 8, 9 and 10 respectively to arrive at U(10) = [5, 5] and V (10) = [4, 6].”
- Federated Residual Learning (2020, background): “In general, fictitious play takes the learner an exponentially long time to converge (Monderer & Shapley, 1996; Brandt et al., 2010).”