The fixed-point bundle method over product-of-simplex domains arising from game equilibria
Abstract
A fixed-point bundle framework extended to product-of-simplex domains via geometric equivalence yields globally convergent path-following algorithms for Nash equilibrium computation.
Developing general-purpose algorithms with global asymptotic convergence guarantees for Nash equilibria of general-sum games remains a longstanding challenge. This paper extends the fixed-point bundle framework for finite-dimensional variational inequalities (VIs) from simplex domains to product-of-simplex domains, enabling its direct application to Nash equilibrium computation. The resulting fixed-point bundle exhibits a composite fiber bundle structure. The key innovation is to construct an equivalent VI on a simplex and establish the equivalence between the two fixed-point bundle frameworks via a fiber bundle isomorphism. Exploiting this geometric equivalence, we show that the path-following algorithm for VIs on product-of-simplex domains inherits the global linear convergence guarantees of the simplex-domain framework. In numerical experiments, the algorithm attained a precision of 10^{-5} on all 5600 randomly generated game instances, with dimension ranging from 2-player 128-action to 128-player 2-action. Code and data are available at https://github.com/shb20tsinghua/FiberBundle_VI.
Community
This paper extends the fiber-bundle path-following framework of arXiv:2606.00778 to product-of-simplex domains, so that the algorithm solves arbitrary m-player n-action general-sum games to arbitrary precision, inheriting global linear convergence: exact for nonsingular equilibria, and up to any prescribed precision for singular ones.
Rather than re-proving convergence, we construct an equivalent variational inequality on a single simplex and establish a fiber bundle isomorphism between the two frameworks, complemented by a slice-extension argument that transfers all guarantees from the original paper. The resulting composite bundle also permits jumps along fibers that avoid singular points, where existing path-following guarantees would otherwise break down.
Experiments on 5600 random instances with dimensions ranging from 2-player 128-action to 128-player 2-action reduce the gap below 1e-5 in every case; code and data are available at https://github.com/shb20tsinghua/FiberBundle_VI.
Get this paper in your agent:
hf papers read 2609.16158 Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper