We present novel techniques for neuro-symbolic concurrent stochastic games, a recently
proposed modelling formalism to represent a set of probabilistic agents operating in a
continuous-space environment using a combination of neural network based perception
mechanisms and traditional symbolic methods. To date, only zero-sum variants of the model
were studied, which is too restrictive when agents have distinct objectives. We formalise
notions of equilibria for these models and present algorithms to synthesise them. Focusing …

Example 1. Consider a two-stage two-agent game with deterministic transitions in Fig. 6, in
which each agent has two actions:{U, D} for agent 1 and {L, R} for agent 2. Non-leaf and leaf
nodes, containing the node numbers, are marked with circles and rectangles, respectively.
For clarity, several histories reaching stage 2 are not displayed here. Edges are labelled
with the associated joint actions. The payoff vectors below leaf nodes are the terminal
rewards, while the payoff vectors below non-leaf nodes denote the unique equilibrium …