Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
144 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Theory and applications of the Sum-Of-Squares technique (2306.16255v3)

Published 28 Jun 2023 in math.OC, cs.IT, math.IT, math.ST, and stat.TH

Abstract: The Sum-of-Squares (SOS) approximation method is a technique used in optimization problems to derive lower bounds on the optimal value of an objective function. By representing the objective function as a sum of squares in a feature space, the SOS method transforms non-convex global optimization problems into solvable semidefinite programs. This note presents an overview of the SOS method. We start with its application in finite-dimensional feature spaces and, subsequently, we extend it to infinite-dimensional feature spaces using reproducing kernels (k-SOS). Additionally, we highlight the utilization of SOS for estimating some relevant quantities in information theory, including the log-partition function.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (23)
  1. J. B. Lasserre, Global optimization with polynomials and the problem of moments, SIAM Journal on Optimization 11, 796 (2001).
  2. P. A. Parrilo, Structured semidefinite programs and semialgebraic geometry methods in robustness and optimization (California Institute of Technology, 2000).
  3. The orthogonality is with respect to the following dot product in matrix space ⟨A,B⟩=tr[A∗⁢B]𝐴𝐵trdelimited-[]superscript𝐴∗𝐵\langle A,B\rangle=\mathop{\rm tr}[A^{\ast}B]⟨ italic_A , italic_B ⟩ = roman_tr [ italic_A start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT italic_B ].
  4. L. Fejér, Über trigonometrische polynome., Journal für die reine und angewandte Mathematik 146, 53 (1916).
  5. U. Grenander and G. Szegö, Toeplitz Forms and Their Applications (University of California Press).
  6. We indicate with 𝔓⁢(⋅)𝔓⋅\mathfrak{P}(\cdot)fraktur_P ( ⋅ ) the power set.
  7. M. Putinar, Positive polynomials on compact semi-algebraic sets, Indiana University Mathematics Journal 42, 969 (1993).
  8. M. X. Goemans and D. P. Williamson, Improved approximation algorithms for maximum cut and satisfiability problems using semidefinite programming, Journal of the ACM (JACM) 42, 1115 (1995).
  9. G. S. Kimeldorf and G. Wahba, A correspondence between bayesian estimation on stochastic processes and smoothing by splines, The Annals of Mathematical Statistics 41, 495 (1970).
  10. N. Aronszajn, Theory of reproducing kernels, Transactions of the American mathematical society 68, 337 (1950).
  11. F. Bach, Learning theory from first principles, Draft of a book, version of Sept 6, 2021 (2021).
  12. U. Marteau-Ferey, F. Bach, and A. Rudi, Non-parametric models for non-negative functions, Advances in Neural Information Processing systems 33, 12816 (2020).
  13. U. Marteau-Ferey, F. Bach, and A. Rudi, Second order conditions to decompose smooth functions as sums of squares, arXiv preprint arXiv:2202.13729  (2022).
  14. A. Rudi, U. Marteau-Ferey, and F. Bach, Finding global minima via kernel approximations, arXiv preprint arXiv:2012.11978  (2020).
  15. P. Del Moral and A. Niclas, A taylor expansion of the square root matrix function, Journal of Mathematical Analysis and Applications 465, 259 (2018).
  16. D. Liberzon, Calculus of variations and optimal control theory: a concise introduction (Princeton University Press, 2011).
  17. R. Vinter, Convex duality and nonlinear optimal control, SIAM Journal on Control and Optimization 31, 518 (1993).
  18. F. Bach, Sum-of-squares relaxations for information theory and variational inference (2022a).
  19. F. Bach, Information theory with kernel methods (2022b).
  20. F. R. Bach and M. I. Jordan, Kernel independent component analysis, Journal of machine learning research 3, 1 (2002).
  21. K. Matsumoto, A new quantum version of f-divergence, in Nagoya Winter Workshop: Reality and Measurement in Algebraic Quantum Theory (Springer, 2015) pp. 229–273.
  22. B. Simon, Orthogonal polynomials on the unit circle. part 1: Classical theory (2005) pp. xxvi+466.
  23. T. M. Cover, Elements of information theory (John Wiley & Sons, 1999).
Citations (2)

Summary

We haven't generated a summary for this paper yet.

X Twitter Logo Streamline Icon: https://streamlinehq.com