Most arbitrage fashions in the present day? Inefficient. Sluggish. Computationally costly.
“Conventional arbitrage strategies counting on convex optimization wrestle in AMM-driven markets as a result of they assume clean constraints that hardly ever exist in follow. As detailed within the paper, convex formulations, whereas efficient for canonical xy=ok curves, fall brief when utilized to concentrated liquidity AMMs.
In these settings, the optimization downside entails piecewise-linear constraints, the place options continuously lie on the boundary of the possible area. This mismatch forces solvers into convergence challenges, as gradient-based strategies fail close to abrupt transitions and simplex strategies bounce between vertices. The result’s a system that, regardless of its mathematical class, is gradual, unpredictable, and ill-suited for real-time arbitrage the place each millisecond counts.”
— Dr. Mark Richardson, Bancor Challenge Lead
