Asymptotic optimality in stochastic optimization (Q2656586)

From MaRDI portal





scientific article
Language Label Description Also known as
default for all languages
No label defined
    English
    Asymptotic optimality in stochastic optimization
    scientific article

      Statements

      Asymptotic optimality in stochastic optimization (English)
      0 references
      0 references
      0 references
      11 March 2021
      0 references
      The authors develop asymptotic theory for stochastic optimization problems. They show a local asymptotic minimax lower bound and make precise connections between tilt stability in optimization and the (statistical) difficulty of solving risk minimization problems. Finally, the authors indicate the best algorithms for solving the given optimization task. A few original theorems are proved. All proofs are correct.
      0 references
      local asymptotic minimax theory
      0 references
      convex analysis
      0 references
      stochastic gradients
      0 references
      manifold identification
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references