Curse of Dimensionality

The curse of dimensionality, a term coined by Richard Bellman in his 1957 book Dynamic Programming, is the failure mode in which the state (and action) space of an MDP grows so large that exact Dynamic Programming — Value Iteration or policy iteration — becomes computationally infeasible. It appears in this vault because it is the binding limitation on the classical MDP route: Pedersen 2023 reports that the dynamic-programming solution of the optimal-execution model “is infeasible for large portfolios,” which is the practical reason reinforcement learning and function approximation are used to find near-optimal policies instead of exact ones.

Connections

• Markov Decision Process Trading Model — suffers_overfitting_risk, source: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4508553
• Dynamic Programming — relates, source: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4508553
• Value Iteration — relates, source: https://en.wikipedia.org/wiki/Richard_E._Bellman
• Richard Bellman — relates, source: https://en.wikipedia.org/wiki/Richard_E._Bellman
• Pedersen 2023 — relates, source: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4508553