Connections between robust statistical estimation, robust decision making with two-stage stochastic optimization, and robust machine learning problems

Ermolieva, T., Ermoliev, Y., Havlik, P. ORCID: https://orcid.org/0000-0001-5551-5085, Derci Augustynczik, A.L., Komendantova, N. ORCID: https://orcid.org/0000-0003-2568-6179, Kahil, T. ORCID: https://orcid.org/0000-0002-7812-5271, Balkovič, J. ORCID: https://orcid.org/0000-0003-2955-4931, Skalský, R. ORCID: https://orcid.org/0000-0002-0983-6897, et al. (2023). Connections between robust statistical estimation, robust decision making with two-stage stochastic optimization, and robust machine learning problems. Cybernetics and systems analysis 59 (3) 33-47.

Full text not available from this repository.

Abstract

The paper discusses connections between the problems of the two-stage stochastic programming, robust decision making, robust statistical estimation and machine learning. In the conditions of uncertainty, possible extreme events and outliers, these problems require quantile-based criteria, constraints, and “goodness-of-fit” indicators. The two-stage STO problems with quantile-based criteria can be effectively solved with the iterative stochastic quasigradient (SQG) solution algorithms. The SQG methods provide a new type of machine learning algorithms that can be effectively used for general type nonsmooth, possibly discontinuous and nonconvex problems, including quantile regression and neural networks training. In general problems of decision-making, feasible solutions, concepts of optimality and robustness are characterized from the context of decision-making situations. Robust ML approaches can be integrated with disciplinary or interdisciplinary decision-making models, e.g., land use, agricultural, energy, etc., for robust decision-making in the conditions of uncertainty, increasing systemic interdependencies and “unknown risks”.

Item Type: Article
Uncontrolled Keywords: two-stage STO, robust decision-making and statistical estimation, robust quantile regression, machine learning, general problems of robust decision making, systemic risks, uncertainties.
Research Programs: Advancing Systems Analysis (ASA)
Advancing Systems Analysis (ASA) > Cooperation and Transformative Governance (CAT)
Advancing Systems Analysis (ASA) > Exploratory Modeling of Human-natural Systems (EM)
Biodiversity and Natural Resources (BNR)
Biodiversity and Natural Resources (BNR) > Agriculture, Forestry, and Ecosystem Services (AFE)
Biodiversity and Natural Resources (BNR) > Integrated Biosphere Futures (IBF)
Biodiversity and Natural Resources (BNR) > Water Security (WAT)
Depositing User: Luke Kirwan
Date Deposited: 30 Mar 2023 11:43
Last Modified: 12 May 2023 05:31
URI: https://pure.iiasa.ac.at/18692

Actions (login required)

View Item View Item