{"id":46,"date":"2022-03-04T17:27:09","date_gmt":"2022-03-04T17:27:09","guid":{"rendered":"https:\/\/web4.bilkent.edu.tr\/saldi\/?page_id=46"},"modified":"2026-03-12T14:52:52","modified_gmt":"2026-03-12T14:52:52","slug":"publications","status":"publish","type":"page","link":"https:\/\/web4.bilkent.edu.tr\/saldi\/?page_id=46","title":{"rendered":"Publications"},"content":{"rendered":"\n

Book<\/strong><\/p>\n\n\n\n

[B1] Naci Saldi, Tamas Linder, and Serdar Yuksel
Finite Approximations in Discrete-Time Stochastic Control: Quantized Models and Asymptotic Optimality<\/a>
<\/strong>Systems & Control: Foundations and Applications Series, Springer, Cham, 2018.<\/p>\n\n\n\n

Journal Papers
<\/strong>
<\/em>Submitted<\/u><\/strong><\/p>\n\n\n\n

[S6] Kagan Akman, Naci Saldi, and Serdar Yuksel.
An Optimal Control Approach To Transformer Training
<\/strong>arXiv<\/a>:<\/a>2603.09571<\/a><\/p>\n\n\n\n

[S5] Sina Sanjari, Naci Saldi, Sinan Gezici, and Serdar Yuksel.
Decentralized Detection with Many Sensors: Optimality of Exchangeable and Identical Encoding Policies
<\/strong>arXiv<\/a>:<\/a>2509.21724<\/a><\/p>\n\n\n\n

[S4] Berkay Anahtarci, Can Deha Kariksiz, and Naci Saldi.
Kernel Based Maximum Entropy Inverse Reinforcement Learning for Mean-Field Games <\/strong><\/strong> arXiv:<\/a>2507.14529<\/a><\/p>\n\n\n\n

[S3] Naci Saldi, Gurdal Arslan, and Serdar Yuksel.
Existence of epsilon-Nash Equilibria in Nonzero-Sum Borel Stochastic Games and Equilibria of Quantized Models
<\/strong><\/strong>arXiv:2411.10805<\/a><\/p>\n\n\n\n

[S2] Naci Saldi.
Best Ergodic Averages via Optimal Graph Filters in Reversible Markov Chains
<\/strong><\/a>arXiv:2405.18995<\/a><\/p>\n\n\n\n

[S1] Sina Sanjari, Naci Saldi, and Serdar Yuksel.
Decentralized Exchangeable Stochastic Dynamic Teams in Coninuous-time, their Mean-Field Limits and Optimality of Symmetric Policies
<\/strong><\/strong>arXiv:2404.15581<\/a><\/p>\n\n\n\n

Published<\/u><\/strong><\/p>\n\n\n\n

[J34] Sina Sanjari, Naci Saldi, and Serdar Yuksel.
Optimality of Decentralized Symmetric Policies for Stochastic Teams with Mean-Field Information Sharing
<\/strong><\/strong>a<\/em><\/em>ccepted to Mathematics of Operations Research, <\/em><\/em>2026.
<\/strong><\/strong>arXiv:2<\/a><\/em><\/em>4<\/a>04.04597<\/a><\/p>\n\n\n\n

[J33] Naci Saldi, Sina Sanjari and Serdar Yuksel.
Quantum Markov Decision Processes: Dynamic and Semi-Definite Programs for Optimal Solutions
<\/strong><\/strong>a<\/em><\/em>ccepted to Applied Mathematics and Optimization, <\/em><\/em>2026.
<\/strong><\/strong>arXiv:2<\/a><\/em><\/em>4<\/a>02.14651<\/a><\/p>\n\n\n\n

[J32] Naci Saldi and Serdar Yuksel.
Kernel Mean Embedding Topology: Weak and Strong Forms for Stochastic Kernels and Implications for Model Learning
<\/strong><\/strong>a<\/em><\/em>ccepted to Annals of Applied Probability, <\/em><\/em>2025.
<\/strong><\/strong>arXiv:2<\/a><\/em><\/em>502.13486<\/a><\/p>\n\n\n\n

[J31] Berkay Anahtarci, Can Deha Kariksiz, and Naci Saldi.
Maximum Causal Entropy IRL in Mean-Field Games and GNEP Framework for Forward RL
<\/strong><\/strong>Journal of Machine Learning Research<\/em>, 26: 1-40, 2025
<\/strong> arXiv:2401.0656<\/a><\/p>\n\n\n\n

[J30] Naci Saldi.
Linear Mean-Field Games with Discounted Cost Criterion
<\/strong><\/strong><\/strong> Mathematics of Operations Research, <\/em><\/em>2025. <\/strong>
<\/strong><\/a>arXiv:2301.06074<\/a><\/p>\n\n\n\n

[J29] U\u011fur Ayd\u0131n, Naci Saldi.
Robustness and Approximation of Discrete-time Mean-field Games under Discounted Cost Criterion
<\/strong><\/strong><\/strong> Mathematics of Operations Research, <\/em><\/em>2025. <\/strong>
<\/strong><\/a>arXiv:2310.10828<\/a><\/p>\n\n\n\n

[J28] Naci Saldi, Sina Sanjari, and Serdar Yuksel.
Quantum Markov Decision Processes: General Theory, Approximations, and Classes of Policies
<\/strong><\/strong>SIAM Journal on Control and Optimization<\/em>, <\/em><\/em>63(2): 1156-1183, 2025.
<\/strong><\/strong>arXiv:2402.14649<\/a><\/em><\/em><\/p>\n\n\n\n

\n

[J27] Naci Saldi.
Common Information Approach for Static Team Problems with Polish Spaces and Existence of Optimal Policies
<\/strong><\/strong>Applied and Computational Mathematics<\/em>,<\/em> 23(3): 307-324, 2024 (Special Issue is dedicated in honor of Tamer Basar<\/em>)<\/em>
<\/strong><\/a>arXiv<\/a>:<\/a>2309.07571<\/a><\/p>\n\n\n\n

<\/p>\n<\/div><\/div>\n\n\n\n

[J26] Sina Sanjari, Naci Saldi, and Serdar Yuksel.
Nash Equilibria for Exchangeable Team against Team Games, their Mean Field Limit, and Role of Common Randomness
<\/strong><\/strong>SIAM Journal on Control and Optimization<\/em>, <\/em>62(3):1437-1464, May 2024
<\/em>arXiv:2210.07339<\/a><\/p>\n\n\n\n

[J25] Ali Devran Kara, Naci Saldi, and Serdar Yuksel.
Q-Learning for MDPs with General Spaces: Convergence and Near Optimality via Quantization under Weak Continuity
<\/strong><\/strong><\/strong>Journal of Machine Learning Research, <\/em>24(199): 1-34, 2023
arXiv:2111.06781<\/a><\/p>\n\n\n\n

[J24] Berkay Anahtarci, Can Deha Kariksiz, and Naci Saldi.
Learning in Discounted-cost and Average-cost Mean-Field Games
<\/strong><\/strong><\/strong>Journal of Machine Learning Research<\/em>, <\/em>24(17): 1-59, 2023
arXiv:1912.13309<\/a><\/p>\n\n\n\n

[J23] Naci Saldi, Tamer Basar, and Maxim Raginsky.
Partially Observed Discrete-time Risk-sensitive Mean Field Games
<\/strong><\/strong><\/strong><\/strong><\/strong>Dynamic Games and Applications, <\/em>13: 929-960, 2023
<\/a><\/a>arXiv<\/a>:2003.11987<\/a><\/p>\n\n\n\n

[J22] Berkay Anahtarci, Can Deha Kariksiz, and Naci Saldi.
Q-learning in Regularized Mean-field Games
<\/strong><\/strong>Dynamic Games and Applications, <\/em>13: 89-117, 2023
<\/a><\/a>arXiv:2003.12151<\/a><\/p>\n\n\n\n

[J21] Naci Saldi and Serdar Yuksel.
Geometry of Information Structures, Strategic Measures and Associated Control Topologies
<\/strong>Probability Surveys, <\/em>19: 450-532, 2022
arXiv:2010.07377<\/a><\/p>\n\n\n\n

[J20] Sina Sanjari, Naci Saldi, and Serdar Yuksel.
Optimality of Independently Randomized Symmetric Policies for Exchangeable Stochastic Teams with Infinitely Many Decision Makers
<\/strong>Mathematics of Operations Research, <\/em>48(3): 1254-1285, August 2023
arXiv:2008.11570<\/a><\/p>\n\n\n\n

[J19] Naci Saldi.
Large Deviations Principle for Discrete-time Mean-field Games
<\/strong>Systems & Control Letters, <\/em>157, November, 2021
arXiv:2102.07865<\/a><\/p>\n\n\n\n

[J18] Naci Saldi.
Regularized Stochastic Team Problems
<\/strong>Systems & Control Letters, <\/em>149, March, 2021
arXiv:2011.03385<\/a><\/p>\n\n\n\n

[J17] Berkay Anahtarci, Can Deha Kariksiz, and Naci Saldi.
Value Iteration Algorithm for Mean-field Games
<\/strong>Systems & Control Letters, <\/em>143, Sep. 2020
arXiv:1909.01758<\/a><\/p>\n\n\n\n

[J16] Naci Saldi, Tamer Basar, and Maxim Raginsky.
Approximate Markov-Nash Equilibria for Discrete-time Risk-sensitive Mean-field Games (link)<\/a>
<\/strong>Mathematics of Operations Research, <\/em>45(4):1596-1620, Nov. 2020<\/p>\n\n\n\n

[J15] Naci Saldi.
Discrete-time average-cost mean-field games on Polish spaces
<\/strong>Turk J Math<\/em>, 44:463-480, 2020
<\/u><\/strong>arXiv:1908.08793<\/a>
<\/u><\/strong><\/u><\/strong>
[J14] Naci Saldi, Serdar Yuksel, and Tamas Linder.
Asymptotic Optimality of Finite Model Approximations for Partially Observed Markov Decision Processes with Discounted Cost<\/strong>
IEEE Trans. Automatic Control<\/em>, 65(1):130-142, Jan. 2020
arXiv:1710.07009<\/a><\/p>\n\n\n\n

[J13] Naci Saldi.
A topology for Team Policies and Existence of Optimal Team Policies in Stochastic Team Theory<\/strong>
IEEE Trans. Automatic Control<\/em>, 65(1):310-317, Jan. 2020
arXiv:1711.00639<\/a><\/p>\n\n\n\n

[J12] Ali Devran Kara, Naci Saldi, and Serdar Yuksel.
Weak Feller Property of Non-linear Filters<\/strong>
<\/u><\/strong>Systems & Control Letters, <\/em>134, Dec. 2019
arXiv:1812.05509<\/a><\/p>\n\n\n\n

[J11] Naci Saldi, Tamer Basar, and Maxim Raginsky.
Approximate Nash Equilibria in Partially Observed Stochastic Games with Mean-Field Interactions
<\/strong>Mathematics of Operations Research, <\/em>44(3):1006-1033, Aug. 2019
arXiv:1705.02036<\/a><\/p>\n\n\n\n

[J10] Naci Saldi.
Finite-State Approximations to Discounted and Average Cost Constrained Markov Decision Processes
<\/strong>IEEE Trans. Automatic Control<\/em>, <\/em><\/strong>64(7): 2681-2696, Jan. 2019
arXiv:1807.02994<\/a><\/p>\n\n\n\n

[J9] Naci Saldi, Tamer Basar, and Maxim Raginsky.
Markov-Nash Equilibria in Mean-Field Games with Discounted Cost<\/strong>
<\/u><\/strong>SIAM Journal on Control and Optimization, <\/em>56(6):4256-4287, Nov. 2018
arXiv:1612.07878<\/a><\/p>\n\n\n\n

<\/u><\/em><\/strong>[J8] Alper Oner, Gulay Oke Gunel and Naci Saldi.
Mean Field Multilayer Stackelberg Differential Games in Multi-Agent Systems<\/strong>
<\/u><\/strong>Applied and Computational Mathematics, <\/em>17(1): 72-95, 2018
<\/u>
<\/strong>[J7] Serdar Yuksel and Naci Saldi.
Convex Analysis in Decentralized Stochastic Control, Strategic Measures and Optimal Solutions<\/strong>
SIAM Journal on Control and Optimization, <\/em>55(1):1-28, Jan. 2017
 arXiv:1609.07685<\/a><\/p>\n\n\n\n

[J6] Naci Saldi, Serdar Yuksel, and Tamas Linder.
Asymptotic Optimality of Finite Approximations to Markov Decision Processes with Borel Spaces<\/strong>
Mathematics of Operations Research<\/em>, 42(4):945-978, Nov. 2017
arXiv:<\/strong>1503.02244<\/a><\/p>\n\n\n\n

[J5] Naci Saldi,Serdar Yuksel, and Tamas Linder.
Finite Model Approximations and Asymptotic Optimality of Quantized Policies in Decentralized Stochastic Control<\/strong>
IEEE Trans. Automatic Control<\/em>, 62(5):2360-2373, May 2017
arXiv:1511.02244<\/a><\/p>\n\n\n\n

[J4] Naci Saldi,Serdar Yuksel, and Tamas Linder.
Near Optimality of Quantized Policies in Stochastic Control Under Weak Continuity Conditions<\/strong>
Journal of Mathematical Analysis and Applications<\/em>, 435(1):321-337, Mar. 2016
arXiv:1410.6985<\/a><\/p>\n\n\n\n

[J3] Naci Saldi, Tamas Linder, and Serdar Yuksel.
Output Constrained Lossy Source Coding with Limited Common Randomness<\/strong>
IEEE Trans. Information Theory<\/em>,  61(9):4984-4998, Sep. 2015
arXiv:1411.5767<\/a><\/p>\n\n\n\n

[J2] Naci Saldi, Tamas Linder, and Serdar Yuksel.
Randomized Quantization and Source Coding  with Constrained Output Distribution<\/strong>
IEEE Trans. Information Theory<\/em>, 61(1):91-106, Jan. 2015
arXiv:1309.2915<\/a><\/p>\n\n\n\n

[J1] Naci Saldi, Tamas Linder, and Serdar Yuksel.
Asymptotic Optimality and Rates of Convergence of Quantized Stationary Policies in Stochastic Control<\/strong>
IEEE Trans. Automatic Control<\/em>, 60(2):553-558, Feb. 2015
arXiv:1310.5770<\/a><\/p>\n\n\n\n

Book Chapters<\/strong><\/p>\n\n\n\n

[BC1] Naci Saldi, Serdar Yuksel, and Tamas Linder
Finite Approximations to Markov Decision Processes with Borel Spaces<\/strong>
In Modern Trends in Controlled Stochastic Processes: Volume II<\/em>, ed: Alexey B. Piunovsky, Luniver Press, 2015<\/p>\n\n\n\n

Conference Papers<\/strong><\/p>\n\n\n\n

[C25] Sina Sanjari, Naci Saldi, Sinan Gezici, and Serdar Yuksel.
Decentralized Detection with Many Sensors: Optimality of Exchangeable and Identical Encoding policies
<\/strong>In Proc. IEEE Conf. Acoustic, Speech, and Signal Processing (ICASSP)<\/em>, 2026<\/p>\n\n\n\n

[C24] Sevket Kaan Alkir and Naci Saldi.
Inverse Reinforcement Learning for Mean-field Games with Average Reward Criterion
<\/strong>In Proc. IEEE Conf. Decision Control (CDC)<\/em>, 2025<\/p>\n\n\n\n

[C23] Naci Saldi and Serdar Yuksel.
Kernel Mean Embedding Topology: Weak and Strong Forms for Stochastic Kernels and Implications for Model Learning
<\/strong>In Proc. IEEE Conf. Decision Control (CDC)<\/em>, 2025<\/p>\n\n\n\n

[C22] Ugur Aydin, Tamer Basar, and Naci Saldi.
Approximation of Discrete-Time Infinite-Horizon Mean-Field   Equilibria via Finite-Horizon Mean-Field Equilibria
<\/strong>In Proc. IEEE Conf. Decision Control (CDC)<\/em>, 2025<\/p>\n\n\n\n

[C21] Naci Saldi, Sina Sanjari, and Serdar Yuksel.
Quantum Markov Decision Processes
<\/strong>In Proc. IEEE Conf. Decision Control (CDC)<\/em>, 2024<\/p>\n\n\n\n

[C20] Sina Sanjari, Naci Saldi, and Serdar Yuksel.
Optimality of Decentralized Symmetric Policies for Stochastic Teams with Mean-Field Information Sharing
<\/strong>In Proc. IEEE Conf. Decision Control (CDC)<\/em>, 2024<\/p>\n\n\n\n

[C19] Sina Sanjari, Naci Saldi, and Serdar Yuksel.
Large Continuous-time Convex Stochastic Teams and Their Mean-Field Limits
<\/strong>In Proc. IEEE American Control Conference (ACC)<\/em>, 2024<\/p>\n\n\n\n

[C18] Sina Sanjari, Naci Saldi, and Serdar Yuksel.
Nash Equilibria for Exchangeable Team against Team Games and their Mean Field Limit
<\/strong>In Proc. IEEE American Control Conference (ACC)<\/em>, 2023<\/p>\n\n\n\n

[C17] Berkay Anahtarci, Can Deha Kariksiz, and Naci Saldi.
Learning in Discrete-time Average-cost Mean-field Games<\/strong>
<\/strong>In Proc. IEEE Conf. Decision Control (CDC)<\/em> 2021<\/p>\n\n\n\n

[C16] Sina Sanjari, Naci Saldi, and Serdar Yuksel.
Independently Randomized Symmetric Policies are Optimal for Exchangeable Stochastic Teams with Infinitely Many Decision Makers<\/strong>
<\/strong>In Proc. IEEE Conf. Decision Control (CDC)<\/em> 2020<\/p>\n\n\n\n

[C15] Naci Saldi, Tamer Basar, and Maxim Raginsky.
Partially-Observed Discrete-Time Risk-Sensitive Mean-Field Games<\/strong>
In Proc. IEEE Conf. Decision Control (CDC)<\/em> 2019 (invited paper)<\/p>\n\n\n\n

[C14] Ali Devran Kara, Naci Saldi, and Serdar Yuksel.
On Weak Feller Continuity Properties of Non-linear Filters<\/strong>
In Proc. IEEE Conf. Decision Control (CDC)<\/em> 2019 (invited paper)<\/p>\n\n\n\n

[C13] Naci Saldi.
A Topology for Policies in Decentralized Stochastic Control and Existence of Optimal Policies<\/strong>
In Proc. IEEE Conf. Decision Control (CDC)<\/em> 2018 (invited paper)<\/p>\n\n\n\n

[C12] Naci Saldi, Tamer Basar, and Maxim Raginsky.
Markov-Nash Equilibria in Mean-Field Games with Discounted Cost<\/strong>
In Proc. IEEE American Control Conference (ACC)<\/em>, 2017<\/p>\n\n\n\n

[C11] Serdar Yuksel and Naci Saldi.
Convex Analysis in Decentralized Stochastic Control and Strategic Measures<\/strong>
In Proc. IEEE Conf. Decision Control (CDC)<\/em> 2016<\/p>\n\n\n\n

[C10] Serdar Yuksel and Naci Saldi.
Strategic Measures Approach to Decentralized Stochastic Control<\/strong>
In Proc. 54th Annual Allerton Conf. Communications, Control, and Computing<\/em>, Monticello, IL, Oct. 2016<\/p>\n\n\n\n

[C9] Naci Saldi, Serdar Yuksel, and Tamas Linder.
Finite Model Approximations and Asymptotic Optimality of Quantized Policies in Decentralized Stochastic Control<\/strong>
In Proc. IEEE Conf. Decision Control (CDC) <\/em>2016<\/p>\n\n\n\n

[C8] Naci Saldi, Tamas Linder, and Serdar Yuksel.
Asymptotic Optimality of Finite Representations and Quantized Policies in Team Problems and Witsenhausen\u2019s Counterexample<\/strong>
In Proc. IEEE American Control Conference (ACC)<\/em>, Boston, Jul. 2016<\/p>\n\n\n\n

[C7] Naci Saldi, Serdar Yuksel, and Tamas Linder.
Finite-State Approximations to Constrained Markov Decision Processes with Borel Spaces<\/strong>
In Proc. 53rd Annual Allerton Conf. Communications, Control, and Computing<\/em>, Monticello, IL, Oct. 2015<\/p>\n\n\n\n

[C6] Naci Saldi, Serdar Yuksel, and Tamas Linder.
Finite-State Approximation of Markov Decision Processes with Unbounded Costs and Borel Spaces<\/strong>
In Proc. IEEE Conf. Decision Control (CDC) 2015<\/em><\/p>\n\n\n\n

[C5] Naci Saldi, Tamas Linder, and Serdar Yuksel.
On Finite State-Action Approximation of Markov Decision Processes with General State and Action Spaces<\/strong>
In Proc. IEEE American Control Conference (ACC)<\/em>, Chicago, Jul. 2015<\/p>\n\n\n\n

[C4] Naci Saldi, Tamas Linder, and Serdar Yuksel.
Asymptotic Optimality of Quantized Policies in Stochastic Control under Weak Continuity Conditions<\/strong>
In Proc. IEEE Conf. Decision Control (CDC)<\/em>, Los Angeles, Dec. 2014<\/p>\n\n\n\n

[C3] Naci Saldi, Tamas Linder, and Serdar Yuksel.
Randomized Source Coding with Limited Common Randomness<\/strong>
In Proc. 52th Annual Allerton Conf. Communications, Control, and Computing<\/em>, Monticello, IL, Oct. 2014<\/p>\n\n\n\n

[C2] Naci Saldi, Tamas Linder, and Serdar Yuksel.
Approximation of Stationary Control Policies by Quantized Control in Markov Decision Processes<\/strong>
In Proc. 51th Annual Allerton Conf. Communications, Control, and Computing<\/em>, Monticello, IL, Oct. 2013<\/p>\n\n\n\n

[C1] Naci Saldi, Tamas Linder, and Serdar Yuksel.
Randomized quantization and optimal design with a marginal constraint<\/strong>
In Proc. IEEE Int. Symp. Inf. Theory (ISIT)<\/em>, Istanbul, Jul. 2013<\/p>\n","protected":false},"excerpt":{"rendered":"

Book [B1] Naci Saldi, Tamas Linder, and Serdar YukselFinite Approximations in Discrete-Time Stochastic Control: Quantized Models and Asymptotic OptimalitySystems & Control: Foundations and Applications Series, Springer, Cham, 2018. Journal PapersSubmitted [S6] Kagan Akman, Naci Saldi, and Serdar Yuksel.An Optimal Control Approach To Transformer TrainingarXiv:2603.09571 [S5] Sina Sanjari, Naci Saldi, Sinan Gezici, and Serdar Yuksel.Decentralized Detection…<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":-1,"comment_status":"closed","ping_status":"closed","template":"","meta":[],"_links":{"self":[{"href":"https:\/\/web4.bilkent.edu.tr\/saldi\/index.php?rest_route=\/wp\/v2\/pages\/46"}],"collection":[{"href":"https:\/\/web4.bilkent.edu.tr\/saldi\/index.php?rest_route=\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/web4.bilkent.edu.tr\/saldi\/index.php?rest_route=\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/web4.bilkent.edu.tr\/saldi\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/web4.bilkent.edu.tr\/saldi\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=46"}],"version-history":[{"count":64,"href":"https:\/\/web4.bilkent.edu.tr\/saldi\/index.php?rest_route=\/wp\/v2\/pages\/46\/revisions"}],"predecessor-version":[{"id":236,"href":"https:\/\/web4.bilkent.edu.tr\/saldi\/index.php?rest_route=\/wp\/v2\/pages\/46\/revisions\/236"}],"wp:attachment":[{"href":"https:\/\/web4.bilkent.edu.tr\/saldi\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=46"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}