Exponential Adaptive Smoothing and Importance Sampling for Optimization of the Conditional Value-at-Risk

Anton Malandii; Boyan Lazarov; Brendan Keith; Stan Uryasev; Will Asness

Exponential Adaptive Smoothing and Importance Sampling for Optimization of the Conditional Value-at-Risk

T0 review · 1 major / 0 minor · reviewed 2026-06-27 · grok-4.3

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The dual representation of CVaR supplies an algorithm with built-in importance sampling that draws from the loss tail.

desk verdict The paper gives a Bregman proximal-point scheme with outer dual updates that adapt the sampling distribution for built-in importance sampling on CVaR, plus a convergence proof for convex cases. read the letter →

arxiv 2606.11515 v1 pith:EPN2NNB6 submitted 2026-06-09 math.OC

Will Asness , Brendan Keith , Boyan Lazarov , Anton Malandii , Stan Uryasev This is my paper · ORCID

classification math.OC

keywords CVaRoptimizationimportancesamplingdualrepresentationstochasticapproximationportfoliosupportvectormachinesBregmanproximalpoint

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper develops a stochastic method for conditional value-at-risk optimization that rests on the dual form of CVaR as a worst-case expectation over a risk envelope. The algorithm alternates inner primal subproblems solved by sampling with outer dual updates that revise the sampling distribution. Because the dual distribution's likelihood ratio converges to the CVaR risk identifier, probability mass automatically concentrates on the tail where only relevant samples lie. The result is an adaptive importance sampler that reduces waste on non-tail outcomes and yields stronger empirical performance than conventional stochastic approximation on portfolio and classification tasks. Convergence holds when the objective is convex.

What carries the argument

The dual probability distribution maintained across outer iterations, which functions as an adaptive importance sampler whose likelihood ratio converges to the CVaR risk identifier.

What would settle it

Execute the algorithm on a convex CVaR problem with known tail and verify that the fraction of samples falling outside the tail declines monotonically while the objective error fails to decrease faster than standard stochastic gradient methods.

Watch

Extended reading notes

Core claim

The method applies the Bregman proximal point algorithm in alternating stochastic primal and dual stages. At each dual stage the probability distribution is updated so that its likelihood ratio relative to the original measure converges to the risk identifier of the CVaR solution. This supplies a built-in importance sampling mechanism that draws with increasing frequency from the tail of the underlying distribution, while samples outside the tail receive vanishing probability. The construction is proved to converge for convex objectives and is demonstrated on portfolio optimization and support-vector machine problems.

Load-bearing premise

The likelihood ratio of the dual probability distributions converges to the risk identifier of the solution's CVaR.

Editorial extensions

If this is right

The algorithm converges for any convex objective function.
Only samples inside the tail contribute meaningfully to the CVaR estimate, while outside-tail samples are drawn with progressively lower probability.
Numerical tests on portfolio optimization and support-vector machine training exhibit better performance than other stochastic approximation schemes.
The dual stage supplies the importance sampling without requiring an external sampler design.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The same dual-update structure could be tested on coherent risk measures other than CVaR to check whether the convergence property carries over.
The adaptive tail focus might reduce variance in related stochastic programs that also depend on extreme outcomes.
If the dual distribution update can be computed in closed form for additional loss families, the method would extend to new application domains without extra sampling overhead.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Desk Editor's Note

read the letter

The core idea is an alternating primal-dual loop where the outer dual stage updates a probability distribution used to sample the inner primal subproblems. The claim is that the likelihood ratio between this dual distribution and the original measure converges to the risk identifier, so samples increasingly focus on the tail that actually matters for CVaR. That is the main novelty relative to plain stochastic approximation.

The paper does deliver a convergence proof for convex objectives and runs the method on portfolio optimization and SVM examples. Those are standard test beds, and having any proof at all is better than most heuristic stochastic CVaR papers.

The soft spot is exactly the one flagged in the stress-test note: the convergence of the likelihood ratio to the risk identifier is stated as a direct consequence of the construction, but the abstract (and the stress-test) gives no regularity conditions, step-size rules, or rate information. If that convergence is only in a weak sense or slower than the primal progress, the promised importance-sampling advantage does not automatically follow. The experiments are described as showing exceptional performance, yet the abstract supplies no baselines, sample sizes, or variance numbers, so it is impossible to judge how much of the gain comes from the dual adaptation versus other implementation choices.

This is a narrow but real contribution inside stochastic optimization for risk measures. Readers working on portfolio problems or robust learning would get value from the algorithm and the proof. It is coherent enough on its own terms to deserve a serious referee rather than a desk reject, even if the convergence details and experimental comparisons will need tightening.

Referee Report

1 major / 0 minor

Summary. The paper introduces a method for CVaR optimization problems using the dual representation of CVaR as the worst-case expectation over a risk envelope. The approach employs the Bregman proximal point algorithm with alternating stochastic primal and dual stages, where the dual stage updates a probability distribution used for sampling in the primal stage. It asserts that the likelihood ratio of these dual distributions converges to the risk identifier, providing built-in importance sampling from the tail of the distribution, which leads to exceptional performance. Convergence is proved for convex objectives, and numerical experiments are presented for portfolio optimization and support vector machines.

Significance. If the convergence of the dual likelihood ratio and the resulting performance improvements are rigorously established, this method could offer a significant advancement in solving CVaR optimization problems efficiently, particularly in applications like financial risk management and machine learning, by reducing variance through adaptive tail sampling without external importance sampling techniques.

major comments (1)

[Abstract] Abstract (paragraph describing the dual stage): The claim that 'the likelihood ratio of the dual probability distributions relative to the distribution underlying the original problem converges to the risk identifier of the solution's CVaR' is asserted without any stated regularity conditions (such as strict convexity of the loss function, compactness of the risk envelope, or restrictions on step sizes). This convergence is load-bearing for the argument that the dual distribution provides a built-in importance sampling mechanism leading to exceptional performance over other stochastic approximation methods.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their careful reading of the manuscript and for highlighting the need for greater precision in the abstract. We respond to the major comment below.

read point-by-point responses

Referee: [Abstract] Abstract (paragraph describing the dual stage): The claim that 'the likelihood ratio of the dual probability distributions relative to the distribution underlying the original problem converges to the risk identifier of the solution's CVaR' is asserted without any stated regularity conditions (such as strict convexity of the loss function, compactness of the risk envelope, or restrictions on step sizes). This convergence is load-bearing for the argument that the dual distribution provides a built-in importance sampling mechanism leading to exceptional performance over other stochastic approximation methods.

Authors: We agree that the abstract should reference the conditions under which the stated convergence holds. The result is proved in Theorem 3.2 for convex objectives, assuming a compact risk envelope and standard step-size conditions (square-summable but not summable). The revised abstract will be updated to read: 'Under standard convexity and step-size assumptions, the likelihood ratio... converges to the risk identifier.' This makes the claim's scope explicit while preserving the abstract's brevity. The full set of assumptions and the proof remain unchanged in Section 3. revision: yes

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; derivation self-contained

full rationale

The paper constructs its algorithm from the dual representation of CVaR (worst-case expectation over risk envelope) and the Bregman proximal point algorithm, with explicit alternation between stochastic primal and dual stages. The convergence of the dual likelihood ratio to the risk identifier is stated as a property of the updates and is accompanied by a separate proof of algorithm convergence for convex objectives. No quoted step reduces a claimed prediction or result to a fitted parameter, self-definition, or self-citation chain by construction. The importance-sampling benefit is derived from the dual-stage mechanism rather than being presupposed. This matches the default expectation for non-circular papers; the provided abstract and structure contain no load-bearing tautologies.

Assumptions & free parameters 0 free parameters · 0 assumptions · 0 invented entities

No free parameters, axioms, or invented entities are identifiable from the abstract; the method is described at the level of an algorithmic template without explicit modeling choices or new postulated objects.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Exponential Adaptive Smoothing and Importance Sampling for Optimization of the Conditional Value-at-Risk." pith.science (2026). https://pith.science/paper/EPN2NNB6

@misc{pith2026260611515,
  author       = {Pith},
  title        = {Pith review of: Exponential Adaptive Smoothing and Importance Sampling for Optimization of the Conditional Value-at-Risk},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EPN2NNB6}},
  note         = {Machine review of arXiv:2606.11515}
}

read the original abstract

We present a novel method for solving conditional value-at-risk (CVaR) optimization problems based on the dual representation of CVaR, which is defined as the worst-case expectation over a risk envelope. The method is based on the Bregman proximal point algorithm and alternates between stochastic primal and dual stages. Every (inner) primal stage involves a subproblem solved by sampling from a probability distribution updated at each dual stage (outer iteration). The likelihood ratio of the dual probability distributions relative to the distribution underlying the original problem converges to the risk identifier of the solution's CVaR. Thus, the dual distribution provides the algorithm with a built-in importance sampling mechanism that draws from the tail of the underlying distribution. Because only samples in the tail influence the CVaR, and samples outside the tail are drawn with decreasing probability, the algorithm delivers exceptional performance over other stochastic approximation methods. We prove the convergence of the algorithm for convex objective functions. Our numerical experiments target representative problems in financial mathematics and machine learning, focusing on portfolio optimization and support-vector machines, respectively.

Figures

Figures reproduced from arXiv: 2606.11515 by the authors.

**Figure 5.** [PITH_FULL_IMAGE:figures/full_fig_p023_5.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

66 extracted references · 2 canonical work pages

[1]

Artzner, F

P. Artzner, F. Delbaen, J-M. Eber, and D. Heath. Coherent Measures of Risk.Mathematical Finance, 9(3):203–228, 1999. 28

1999
[2]

Computing VaR and CVaR using stochastic approximation and adaptive unconstrained importance sampling.Monte Carlo Methods and Applications, 15(3):173–210, 2009

Olivier Bardou, Noufel Frikha, and Gilles Pag‘es. Computing VaR and CVaR using stochastic approximation and adaptive unconstrained importance sampling.Monte Carlo Methods and Applications, 15(3):173–210, 2009

2009
[3]

Adaptive sampling strategies for risk-averse stochastic optimization with constraints.IMA Journal of Numerical Analysis, 43(6):3729–3765, 2023

Florian Beiser, Brendan Keith, Simon Urbainczyk, and Barbara Wohlmuth. Adaptive sampling strategies for risk-averse stochastic optimization with constraints.IMA Journal of Numerical Analysis, 43(6):3729–3765, 2023

2023
[4]

Bertsekas.Nonlinear Programming

Dimitri P. Bertsekas.Nonlinear Programming. Athena Scientific, Belmont, MA, 2 edition, 1999

1999
[5]

L. M. Bregman. The Relaxation Method of Finding the Common Point of Convex Sets and Its Application to the Solution of Problems in Convex Programming.USSR Computational Mathematics and Mathematical Physics, 7(3):200–217, 1967

1967
[6]

D. Cajas. Riskfolio-Lib (6.2.0), 2024

2024
[7]

Maximum Block Improvement and Polynomial Optimization

Bo Chen, Shaohua He, Zhiping Li, and Shuzhong Zhang. Maximum Block Improvement and Polynomial Optimization. SIAM Journal on Optimization, 22(1):87–107, 2012

2012
[8]

Chen and M

G. Chen and M. Teboulle. Convergence Analysis of a Proximal-Like Minimization Algorithm Using Bregman Functions. SIAM Journal on Optimization, 3(3):538–543, 1993

1993

Show all 66 references

[9]

Csisz´ ar

I. Csisz´ ar. Why Least Squares and Maximum Entropy? An Axiomatic Approach to Inference for Linear Inverse Problems. Annals of Statistics, 19(4):2032–2066, 1991
[10]

Csisz´ ar

I. Csisz´ ar. Maximum Entropy and Related Methods. InTrans. 12th Prague Conf. Information Theory, Statistical Decision Functions and Random Processes, pages 58–62, Prague, Czech Acad. Sci., 1994

1994
[11]

Csisz´ ar

I. Csisz´ ar. Generalized Projections for Non-Negative Functions.Acta Mathematica Hungarica, 68:161–186, 1995

1995
[12]

Csisz´ ar and F

I. Csisz´ ar and F. Mat´ uˇ s. On Minimization of Entropy Functionals under Moment Constraints. InProceedings of ISIT 2008, pages 2101–2105, Toronto, Canada, 2008

2008
[13]

Csisz´ ar and F

I. Csisz´ ar and F. Mat´ uˇ s. On Minimization of Multivariate Entropy Functionals. InProceedings of ITW 2009, pages 96–100, Volos, Greece, 2009

2009
[14]

Danskin.The Theory of Max-Min and Its Application to Weapons Allocation Problems, volume 5 ofEconometrics and Operations Research

John M. Danskin.The Theory of Max-Min and Its Application to Weapons Allocation Problems, volume 5 ofEconometrics and Operations Research. Springer, Berlin, Heidelberg, 1967

1967
[15]

Efficient black-box importance sampling for VaR and CVaR estimation

Anand Deo and Karthyek Murthy. Efficient black-box importance sampling for VaR and CVaR estimation. InProceedings of the 2021 Winter Simulation Conference (WSC), pages 1–12. IEEE, 2021

2021
[16]

Approximate Iterations in Bregman-Function-Based Proximal Algorithms.Mathematical program- ming, 83(1):113–123, 1998

Jonathan Eckstein. Approximate Iterations in Bregman-Function-Based Proximal Algorithms.Mathematical program- ming, 83(1):113–123, 1998

1998
[17]

Grechuk, A

B. Grechuk, A. Malandii, R. T. Rockafellar, and S. Uryasev. The Risk Quadrangle in Optimization: An Overview with Recent Results and Extensions.EURO Journal on Computational Optimization, 14:100129, 2026

2026
[18]

Shengyi He, Guangxin Jiang, Henry Lam, and Michael C. Fu. Adaptive importance sampling for efficient stochastic root finding and quantile estimation.Operations Research, 72(6):2612–2630, 2024

2024
[19]

Risk-averse design of tall buildings for uncertain wind conditions.Computer Methods in Applied Mechanics and Engineering, 402:115371, 2022

Anoop Kodakkal, Brendan Keith, Ustim Khristenko, Andreas Apostolatos, Kai-Uwe Bletzinger, Barbara Wohlmuth, and Roland W¨ uchner. Risk-averse design of tall buildings for uncertain wind conditions.Computer Methods in Applied Mechanics and Engineering, 402:115371, 2022

2022
[20]

D. P. Kouri and T. M. Surowiec. Epi-Regularization of Risk Measures.Mathematics of Operations Research, 45(2):774– 795, 2020

2020
[21]

D. P. Kouri and T. M. Surowiec. A Primal–Dual Algorithm for Risk Minimization.Mathematical Programming, 193:337– 363, 2022

2022
[22]

All Roads Lead to Rome: Path-Following Augmented Lagrangian Methods via Bregman Proximal Regularization.arXiv preprint arXiv.2602.15710, 2026

Emanuel Laude. All Roads Lead to Rome: Path-Following Augmented Lagrangian Methods via Bregman Proximal Regularization.arXiv preprint arXiv.2602.15710, 2026

2026
[23]

Dennis Leventhal and Adrian S. Lewis. Randomized Methods for Linear Constraints: Convergence Rates and Conditioning. Mathematics of Operations Research, 35(3):641–654, 2010

2010
[24]

On the Convergence of the Coordinate Descent Method for Convex Differentiable Mini- mization.Journal of Optimization Theory and Applications, 72(1):7–35, 1992

Zhi-Quan Luo and Paul Tseng. On the Convergence of the Coordinate Descent Method for Convex Differentiable Mini- mization.Journal of Optimization Theory and Applications, 72(1):7–35, 1992

1992
[25]

Adaptive Gradient Descent Without Descent.arXiv preprint arXiv:1910.09529, 2019

Yura Malitsky and Konstantin Mishchenko. Adaptive Gradient Descent Without Descent.arXiv preprint arXiv:1910.09529, 2019

1910
[26]

Adaptive Proximal Gradient Method for Convex Optimization.Advances in Neural Information Processing Systems, 37:100670–100697, 2024

Yura Malitsky and Konstantin Mishchenko. Adaptive Proximal Gradient Method for Convex Optimization.Advances in Neural Information Processing Systems, 37:100670–100697, 2024

2024
[27]

Martinet

B. Martinet. R´ egularisation d'in´ equations variationnelles par approximations successives.Revue Fran¸ caise d’Informatique et de Recherche Op´ erationnelle. S´ erie Rouge, 4(R–3):154–158, 1970

1970
[28]

CVaR Portfolio Optimization

MathWorks Quant Team. CVaR Portfolio Optimization. MATLAB Central File Exchange, 2024. Retrieved July 12, 2024

2024
[29]

Mirzoakhmedov and S

F. Mirzoakhmedov and S. P. Uryasev. Adaptive Step Adjustment for a Stochastic Optimization Algorithm.USSR Computational Mathematics and Mathematical Physics, 23(6):20–27, 1983

1983
[30]

J.-J. Moreau. Proximit´ e et dualit´ e dans un espace hilbertien.Bulletin de la Soci´ et´ e Math´ ematique de France, 93:273–299, 1965

1965
[31]

Nemirovski and D

A. Nemirovski and D. Yudin.Problem Complexity and Method Efficiency in Optimization. Wiley, New York, 1983

1983
[32]

Efficiency of Coordinate Descent Methods on Huge-Scale Optimization Problems.SIAM Journal on Optimization, 22(2):341–362, 2012

Yurii Nesterov. Efficiency of Coordinate Descent Methods on Huge-Scale Optimization Problems.SIAM Journal on Optimization, 22(2):341–362, 2012

2012
[33]

Springer, 2018

Yurii Nesterov.Lectures on Convex Optimization, volume 137. Springer, 2018

2018
[34]

Efficient Random Coordinate Descent Algorithms for Large-Scale Structured Nonconvex Optimization.Journal of Global Optimization, 61(1):19–46, 2015

Adrian Patrascu and Ion Necoara. Efficient Random Coordinate Descent Algorithms for Large-Scale Structured Nonconvex Optimization.Journal of Global Optimization, 61(1):19–46, 2015

2015
[35]

E. Pauwels. On the Nature of Bregman Functions.Operations Research Letters, 57:107183, 2024

2024
[36]

An adaptive importance sampling algorithm for risk-averse optimization.Journal of Computational Physics, 547:114548, 2026

Sandra Pieraccini and Tommaso Vanzan. An adaptive importance sampling algorithm for risk-averse optimization.Journal of Computational Physics, 547:114548, 2026

2026
[37]

Pinsker.Information and Information Stability of Random Variables and Processes

Mark S. Pinsker.Information and Information Stability of Random Variables and Processes. Holden-Day, San Francisco,
[38]

Translated and edited by Amiel Feinstein
[39]

Portfolio Safeguard Help Manual, 2026

Portfolio Safeguard. Portfolio Safeguard Help Manual, 2026. (accessed: 2026-3-4)

2026
[40]

L. A. Prashanth. Policy gradients for CVaR-constrained MDPs. InAlgorithmic Learning Theory, volume 8776 ofLecture 29 Notes in Computer Science, pages 155–169, Cham, 2014. Springer

2014
[41]

Iteration Complexity of Randomized Block-Coordinate Descent Methods for Minimizing a Composite Function.Mathematical Programming, 144(1–2):1–38, 2014

Peter Richt´ arik and Martin Tak´ aˇ c. Iteration Complexity of Randomized Block-Coordinate Descent Methods for Minimizing a Composite Function.Mathematical Programming, 144(1–2):1–38, 2014

2014
[42]

On Optimal Probabilities in Stochastic Coordinate Descent Methods.Optimization Letters, 10(6):1233–1243, 2016

Peter Richt´ arik and Martin Tak´ aˇ c. On Optimal Probabilities in Stochastic Coordinate Descent Methods.Optimization Letters, 10(6):1233–1243, 2016

2016
[43]

Robbins and S

H. Robbins and S. Monro. A Stochastic Approximation Method.Annals of Mathematical Statistics, 22(3):400–407, September 1951

1951
[44]

A Convergence Theorem for Nonnegative Almost Supermartingales and Some Applications

Herbert Robbins and David Siegmund. A Convergence Theorem for Nonnegative Almost Supermartingales and Some Applications. In T. L. Lai and D. Siegmund, editors,Herbert Robbins Selected Papers, pages 111–135. Springer, 1985

1985
[45]

R. T. Rockafellar. Monotone Operators and the Proximal Point Algorithm.SIAM Journal on Control and Optimization, 14(5):877–898, 1976

1976
[46]

R. T. Rockafellar and J. O. Royset. Engineering Decisions under Risk Averseness.ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part A: Civil Engineering, 1(2):04015003, 2015

2015
[47]

R. T. Rockafellar and S. Uryasev. Optimization of Conditional Value-at-Risk.Journal of Risk, 2:21–42, 2000

2000
[48]

R. T. Rockafellar and S. Uryasev. The Fundamental Risk Quadrangle in Risk Management, Optimization and Statistical Estimation.Surveys in Operations Research and Management Science, 18(1-2):33–53, oct 2013

2013
[49]

R. T. Rockafellar, S. Uryasev, and M. Zabarankin. Generalized Deviations in Risk Analysis.Finance and Stochastics, 10(1):51–74, 2006

2006
[50]

Augmented Lagrangians and Applications of the Proximal Point Algorithm in Convex Programming

R Tyrrell Rockafellar. Augmented Lagrangians and Applications of the Proximal Point Algorithm in Convex Programming. Mathematics of operations research, 1(2):97–116, 1976

1976
[51]

J. O. Royset. Risk-Adaptive Approaches to Stochastic Optimization: A Survey.SIAM Review, 67(1):3–70, 2025

2025
[52]

J. O. Royset and R. J-B. Wets.An Optimization Primer. Springer, 2021

2021
[53]

E. K. Ryu and W. Yin.Large-Scale Convex Optimization: Algorithms & Analyses via Monotone Operators. Cambridge University Press, 2022

2022
[54]

Stochastic Methods forℓ 1 Regularized Loss Minimization

Shai Shalev-Shwartz and Ambuj Tewari. Stochastic Methods forℓ 1 Regularized Loss Minimization. InProceedings of the 26th Annual International Conference on Machine Learning (ICML), pages 929–936, Montreal, Quebec, Canada,
[55]

Stochastic Methods forℓ 1 Regularized Loss Minimization.Journal of Machine Learning Research, 12(52):1865–1892, 2011

Shai Shalev-Shwartz and Ambuj Tewari. Stochastic Methods forℓ 1 Regularized Loss Minimization.Journal of Machine Learning Research, 12(52):1865–1892, 2011

2011
[56]

Shor.Nondifferentiable Optimization and Polynomial Problems, volume 24 ofNonconvex Optimization and Its Applications

Naum Z. Shor.Nondifferentiable Optimization and Polynomial Problems, volume 24 ofNonconvex Optimization and Its Applications. Springer, New York, NY, 1 edition, 1998

1998
[57]

An Inexact Hybrid Generalized Proximal Point Algorithm and Some New Results on the Theory of Bregman Functions.Mathematics of Operations Research, 25(2):214–230, 2000

Mikhail V Solodov and Benar Fux Svaiter. An Inexact Hybrid Generalized Proximal Point Algorithm and Some New Results on the Theory of Bregman Functions.Mathematics of Operations Research, 25(2):214–230, 2000

2000
[58]

P. I. Stetsyuk. Theory and Software Implementations of Shor’s r-Algorithms.Cybernetics and Systems Analysis, 53(5):692– 703, 2017

2017
[59]

InProceed- ings of the 25th International Conference on Machine Learning, ICML ’08, page 1056–1063, New York, NY, USA,

Akiko Takeda and Masashi Sugiyama.ν-Support Vector Machine as Conditional Value-at-Risk Minimization. InProceed- ings of the 25th International Conference on Machine Learning, ICML ’08, page 1056–1063, New York, NY, USA,
[60]

Association for Computing Machinery
[61]

Optimizing the CVaR via sampling

Aviv Tamar, Yonatan Glassner, and Shie Mannor. Optimizing the CVaR via sampling. InProceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence, pages 2993–2999. AAAI Press, 2015

2015
[62]

Teboulle

M. Teboulle. A Simplified View of First-Order Methods for Optimization.Mathematical Programming, 170(1):67–96, 2018

2018
[63]

Dual Ascent Methods for Problems with Strictly Convex Costs and Linear Constraints: A Unified Approach

Paul Tseng. Dual Ascent Methods for Problems with Strictly Convex Costs and Linear Constraints: A Unified Approach. SIAM Journal on Control and Optimization, 28(1):214–242, 1990

1990
[64]

Tsybakov.Introduction to Nonparametric Estimation

Alexandre B. Tsybakov.Introduction to Nonparametric Estimation. Springer Series in Statistics. Springer, New York, 2009

2009
[65]

Solving Large-Scale Linear Prediction Problems Using Stochastic Gradient Descent Algorithms

Tong Zhang. Solving Large-Scale Linear Prediction Problems Using Stochastic Gradient Descent Algorithms. InProceed- ings of the 21st International Conference on Machine Learning (ICML). ACM, 2004

2004
[66]

Stochastic Primal–Dual Coordinate Method for Regularized Empirical Risk Minimization

Yuchen Zhang and Lin Xiao. Stochastic Primal–Dual Coordinate Method for Regularized Empirical Risk Minimization. InProceedings of the 32nd International Conference on Machine Learning (ICML), pages 353–361. JMLR Workshop and Conference Proceedings, 2015. 30

2015

Pith tools

Reviewed June 27, 2026 · model on record in the stance chip above.

[1] [1]

Artzner, F

P. Artzner, F. Delbaen, J-M. Eber, and D. Heath. Coherent Measures of Risk.Mathematical Finance, 9(3):203–228, 1999. 28

1999

[2] [2]

Computing VaR and CVaR using stochastic approximation and adaptive unconstrained importance sampling.Monte Carlo Methods and Applications, 15(3):173–210, 2009

Olivier Bardou, Noufel Frikha, and Gilles Pag‘es. Computing VaR and CVaR using stochastic approximation and adaptive unconstrained importance sampling.Monte Carlo Methods and Applications, 15(3):173–210, 2009

2009

[3] [3]

Adaptive sampling strategies for risk-averse stochastic optimization with constraints.IMA Journal of Numerical Analysis, 43(6):3729–3765, 2023

Florian Beiser, Brendan Keith, Simon Urbainczyk, and Barbara Wohlmuth. Adaptive sampling strategies for risk-averse stochastic optimization with constraints.IMA Journal of Numerical Analysis, 43(6):3729–3765, 2023

2023

[4] [4]

Bertsekas.Nonlinear Programming

Dimitri P. Bertsekas.Nonlinear Programming. Athena Scientific, Belmont, MA, 2 edition, 1999

1999

[5] [5]

L. M. Bregman. The Relaxation Method of Finding the Common Point of Convex Sets and Its Application to the Solution of Problems in Convex Programming.USSR Computational Mathematics and Mathematical Physics, 7(3):200–217, 1967

1967

[6] [6]

D. Cajas. Riskfolio-Lib (6.2.0), 2024

2024

[7] [7]

Maximum Block Improvement and Polynomial Optimization

Bo Chen, Shaohua He, Zhiping Li, and Shuzhong Zhang. Maximum Block Improvement and Polynomial Optimization. SIAM Journal on Optimization, 22(1):87–107, 2012

2012

[8] [8]

Chen and M

G. Chen and M. Teboulle. Convergence Analysis of a Proximal-Like Minimization Algorithm Using Bregman Functions. SIAM Journal on Optimization, 3(3):538–543, 1993

1993