Eitan Altman
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Eitan Altman has a total of 268 co-authors. Here are the top co-authors by number of shared publications:
- Rachid El-Azouzi
Avignon University, Avignon, France
17 shared publications - Yezekael Hayel
University of Avignon
10 shared publications - Zwi Altman
Orange Labs 38/40 rue du Général Leclerc 92794 Issy-les-Moulineaux
10 shared publications Corinne Touati
Univrsity of Grenobles
6 shared publications- Konstantin Avrachenkov
Inria, Université Côte d’Azur
6 shared publications - Samir Medina Perlaza
INRIA, Centre Inria dUniversit Cte dAzur, Sophia Antipolis, France
6 shared publications
Latest Publications
- 2 January 2026
- Published by Springer Nature
- p. 26-46
Games Among Selfish and Team Stations in Polling Systems
- 2 January 2026
- Published by Springer Nature
- p. 26-46
Games Among Selfish and Team Stations in Polling Systems
- 5 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 70 (11) , 7125-7140
Learning Market Equilibria Preserving Statistical Privacy Using Performative Prediction
We consider a peer-to-peer electricity market modeled as a private network game, where end users minimize their cost by computing their demand and controllable generation. Their nominal demand constitutes sensitive information that they might want to keep private. We prove that the private network game admits a unique variational equilibrium, which depends on the private information of all end users. Thus, to update their strategy, end users rely on randomized readings. A data aggregator is introduced, which aims to learn the end users' private information, while remunerating them depending on the quality of their readings. Using performative prediction, we define a decision-dependent game explicitly taking into account the distribution shift caused by the end users' hidden ability. The decision-dependent game coincides with a Stackelberg game when the end users' hidden abilities are best responses. Further, the market robustness can be quantified by evaluating the efficiency loss as the difference between the social cost in the performatively stable equilibrium and the optimum. We show that under mild assumptions, the performatively stable equilibrium can be found by distributed and sequential variants of the repeated stochastic gradient method while we propose a two-timescale stochastic approximation method to learn Stackelberg equilibrium. Finally, we formulate the data aggregator's optimal contract design as a bilevel optimization problem that we cast as a more tractable nonlinear nonconvex optimization problem, which can be solved using simulated annealing. Simulations on small and large scale problem instances illustrate the results.
- 5 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 70 (11) , 7125-7140
Learning Market Equilibria Preserving Statistical Privacy Using Performative Prediction
We consider a peer-to-peer electricity market modeled as a private network game, where end users minimize their cost by computing their demand and controllable generation. Their nominal demand constitutes sensitive information that they might want to keep private. We prove that the private network game admits a unique variational equilibrium, which depends on the private information of all end users. Thus, to update their strategy, end users rely on randomized readings. A data aggregator is introduced, which aims to learn the end users' private information, while remunerating them depending on the quality of their readings. Using performative prediction, we define a decision-dependent game explicitly taking into account the distribution shift caused by the end users' hidden ability. The decision-dependent game coincides with a Stackelberg game when the end users' hidden abilities are best responses. Further, the market robustness can be quantified by evaluating the efficiency loss as the difference between the social cost in the performatively stable equilibrium and the optimum. We show that under mild assumptions, the performatively stable equilibrium can be found by distributed and sequential variants of the repeated stochastic gradient method while we propose a two-timescale stochastic approximation method to learn Stackelberg equilibrium. Finally, we formulate the data aggregator's optimal contract design as a bilevel optimization problem that we cast as a more tractable nonlinear nonconvex optimization problem, which can be solved using simulated annealing. Simulations on small and large scale problem instances illustrate the results.
- 1 April 2025
- Published by Springer Nature in Queueing Systems
- Vol. 109 (2) , 1-34
Queues with inspection cost: To see or not to see?
- 1 April 2025
- Published by Springer Nature in Queueing Systems
- Vol. 109 (2) , 1-34
Queues with inspection cost: To see or not to see?
- Open Access
Queues with inspection cost: To see or not to see?
Consider an M/M/1-type queue where joining attains a known reward, but a known waiting cost is paid per time unit spent queueing. In the 1960s, Naor showed that any arrival optimally joins the queue if its length is less than a known threshold. Yet acquiring knowledge of the queue length often brings an additional cost, e.g., website loading time or data roaming charge. Therefore, our model presents any arrival with three options: join blindly, balk blindly, or pay a known inspection cost to make the optimal joining decision by comparing the queue length to Naor's threshold. In a recent paper, Hassin and Roet-Green prove that a unique Nash equilibrium always exists and classify regions where the equilibrium probabilities are non-zero. We complement these findings with new closed-form expressions for the equilibrium probabilities in the majority of cases. Further, Hassin and Roet-Green show that minimizing inspection cost maximises social welfare. Envisaging a queue operator choosing where to invest, we compare the effects of lowering inspection cost and increasing the queue-joining reward on social welfare. We prove that the former dominates and that the latter can even have a detrimental effect on social welfare.
Open AccessQueues with inspection cost: To see or not to see?
Consider an M/M/1-type queue where joining attains a known reward, but a known waiting cost is paid per time unit spent queueing. In the 1960s, Naor showed that any arrival optimally joins the queue if its length is less than a known threshold. Yet acquiring knowledge of the queue length often brings an additional cost, e.g., website loading time or data roaming charge. Therefore, our model presents any arrival with three options: join blindly, balk blindly, or pay a known inspection cost to make the optimal joining decision by comparing the queue length to Naor's threshold. In a recent paper, Hassin and Roet-Green prove that a unique Nash equilibrium always exists and classify regions where the equilibrium probabilities are non-zero. We complement these findings with new closed-form expressions for the equilibrium probabilities in the majority of cases. Further, Hassin and Roet-Green show that minimizing inspection cost maximises social welfare. Envisaging a queue operator choosing where to invest, we compare the effects of lowering inspection cost and increasing the queue-joining reward on social welfare. We prove that the former dominates and that the latter can even have a detrimental effect on social welfare.
- 30 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 23 (10) , 13752-13765
Design and Cross-Layer Optimization of Low Cost RIS-Assisted Communication Systems
The deployment of RISs in future 6G networks is expected to substantially improve mobile network coverage. This paper introduces a new cross-layer low-complexity scheme for the online optimization of RIS-assisted communication systems. It jointly combines BS and RIS configuration and fair UEs’ scheduling which is critical for high-performance deployments. A RIS beam synthesis method is especially proposed for RIS configuration. The proposed solution embeds two nested control loops: i) a fast control loop working at the OFDMA slot scale and consisting in a standard UEs proportional fair scheduler, and ii) a slow control loop operating at the OFDMA frame scale which adapts the RIS’ configuration to the UEs’ spatial distribution and maximizes the UEs’ aggregated performance. The slow control loop is based on an online stochastic approximation algorithm whose convergence to the optimal restpoint is proved. In a reference scenario, the proposed scheduler achieves a gain of 47% in mean spectral efficiency for NLOS UEs over a baseline scheme.
- 30 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 23 (10) , 13752-13765
Design and Cross-Layer Optimization of Low Cost RIS-Assisted Communication Systems
The deployment of RISs in future 6G networks is expected to substantially improve mobile network coverage. This paper introduces a new cross-layer low-complexity scheme for the online optimization of RIS-assisted communication systems. It jointly combines BS and RIS configuration and fair UEs’ scheduling which is critical for high-performance deployments. A RIS beam synthesis method is especially proposed for RIS configuration. The proposed solution embeds two nested control loops: i) a fast control loop working at the OFDMA slot scale and consisting in a standard UEs proportional fair scheduler, and ii) a slow control loop operating at the OFDMA frame scale which adapts the RIS’ configuration to the UEs’ spatial distribution and maximizes the UEs’ aggregated performance. The slow control loop is based on an online stochastic approximation algorithm whose convergence to the optimal restpoint is proved. In a reference scenario, the proposed scheduler achieves a gain of 47% in mean spectral efficiency for NLOS UEs over a baseline scheme.
- 2 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Information Theory
- Vol. 5, 175-189
The Worst-Case Data-Generating Probability Measure in Statistical Learning
The worst-case data-generating (WCDG) probability measure is introduced as a tool for characterizing the generalization capabilities of machine learning algorithms. Such a WCDG probability measure is shown to be the unique solution to two different optimization problems: (a) The maximization of the expected loss over the set of probability measures on the datasets whose relative entropy with respect to a reference measure is not larger than a given threshold; and (b) The maximization of the expected loss with regularization by relative entropy with respect to the reference measure. Such a reference measure can be interpreted as a prior on the datasets. The WCDG cumulants are finite and bounded in terms of the cumulants of the reference measure. To analyze the concentration of the expected empirical induced by the WCDG probability measure, the notion of (,δ)-robustness of models is introduced. Closed-form expressions are presented for the sensitivity of the expected loss for a fixed model. These tools result in the characterization of a novel expression for the generalization error of arbitrary machine learning algorithms. This exact expression is provided in terms of the WCDG probability measure and leads to an upper bound that is equal to the sum of the mutual information and the lautum information between the models and the datasets, up to a constant factor. This upper bound is achieved by a Gibbs algorithm. This finding reveals that an exploration into the generalization error of the Gibbs algorithm facilitates the derivation of overarching insights applicable to any machine learning algorithm.
- 2 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Information Theory
- Vol. 5, 175-189
The Worst-Case Data-Generating Probability Measure in Statistical Learning
The worst-case data-generating (WCDG) probability measure is introduced as a tool for characterizing the generalization capabilities of machine learning algorithms. Such a WCDG probability measure is shown to be the unique solution to two different optimization problems: (a) The maximization of the expected loss over the set of probability measures on the datasets whose relative entropy with respect to a reference measure is not larger than a given threshold; and (b) The maximization of the expected loss with regularization by relative entropy with respect to the reference measure. Such a reference measure can be interpreted as a prior on the datasets. The WCDG cumulants are finite and bounded in terms of the cumulants of the reference measure. To analyze the concentration of the expected empirical induced by the WCDG probability measure, the notion of (,δ)-robustness of models is introduced. Closed-form expressions are presented for the sensitivity of the expected loss for a fixed model. These tools result in the characterization of a novel expression for the generalization error of arbitrary machine learning algorithms. This exact expression is provided in terms of the WCDG probability measure and leads to an upper bound that is equal to the sum of the mutual information and the lautum information between the models and the datasets, up to a constant factor. This upper bound is achieved by a Gibbs algorithm. This finding reveals that an exploration into the generalization error of the Gibbs algorithm facilitates the derivation of overarching insights applicable to any machine learning algorithm.
- Open Access
- 24 March 2024
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 38 (15) , 17271-17279
Generalization Analysis of Machine Learning Algorithms via the Worst-Case Data-Generating Probability Measure
In this paper, the worst-case probability measure over the data is introduced as a tool for characterizing the generalization capabilities of machine learning algorithms. More specifically, the worst-case probability measure is a Gibbs probability measure and the unique solution to the maximization of the expected loss under a relative entropy constraint with respect to a reference probability measure. Fundamental generalization metrics, such as the sensitivity of the expected loss, the sensitivity of the empirical risk, and the generalization gap are shown to have closed-form expressions involving the worst-case data-generating probability measure. Existing results for the Gibbs algorithm, such as characterizing the generalization gap as a sum of mutual information and lautum information, up to a constant factor, are recovered. A novel parallel is established between the worst-case data-generating probability measure and the Gibbs algorithm. Specifically, the Gibbs probability measure is identified as a fundamental commonality of the model space and the data space for machine learning algorithms.
- 24 March 2024
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 38 (15) , 17271-17279
Open AccessGeneralization Analysis of Machine Learning Algorithms via the Worst-Case Data-Generating Probability Measure
In this paper, the worst-case probability measure over the data is introduced as a tool for characterizing the generalization capabilities of machine learning algorithms. More specifically, the worst-case probability measure is a Gibbs probability measure and the unique solution to the maximization of the expected loss under a relative entropy constraint with respect to a reference probability measure. Fundamental generalization metrics, such as the sensitivity of the expected loss, the sensitivity of the empirical risk, and the generalization gap are shown to have closed-form expressions involving the worst-case data-generating probability measure. Existing results for the Gibbs algorithm, such as characterizing the generalization gap as a sum of mutual information and lautum information, up to a constant factor, are recovered. A novel parallel is established between the worst-case data-generating probability measure and the Gibbs algorithm. Specifically, the Gibbs probability measure is identified as a fundamental commonality of the model space and the data space for machine learning algorithms.
- Open Access
Generalization Analysis of Machine Learning Algorithms via the Worst-Case Data-Generating Probability Measure
In this paper, the worst-case probability measure over the data is introduced as a tool for characterizing the generalization capabilities of machine learning algorithms. More specifically, the worst-case probability measure is a Gibbs probability measure and the unique solution to the maximization of the expected loss under a relative entropy constraint with respect to a reference probability measure. Fundamental generalization metrics, such as the sensitivity of the expected loss, the sensitivity of the empirical risk, and the generalization gap are shown to have closed-form expressions involving the worst-case data-generating probability measure. Existing results for the Gibbs algorithm, such as characterizing the generalization gap as a sum of mutual information and lautum information, up to a constant factor, are recovered. A novel parallel is established between the worst-case data-generating probability measure and the Gibbs algorithm. Specifically, the Gibbs probability measure is identified as a fundamental commonality of the model space and the data space for machine learning algorithms.
Open AccessGeneralization Analysis of Machine Learning Algorithms via the Worst-Case Data-Generating Probability Measure
In this paper, the worst-case probability measure over the data is introduced as a tool for characterizing the generalization capabilities of machine learning algorithms. More specifically, the worst-case probability measure is a Gibbs probability measure and the unique solution to the maximization of the expected loss under a relative entropy constraint with respect to a reference probability measure. Fundamental generalization metrics, such as the sensitivity of the expected loss, the sensitivity of the empirical risk, and the generalization gap are shown to have closed-form expressions involving the worst-case data-generating probability measure. Existing results for the Gibbs algorithm, such as characterizing the generalization gap as a sum of mutual information and lautum information, up to a constant factor, are recovered. A novel parallel is established between the worst-case data-generating probability measure and the Gibbs algorithm. Specifically, the Gibbs probability measure is identified as a fundamental commonality of the model space and the data space for machine learning algorithms.
- Open Access
- 30 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 83-92
Learning Optimal Edge Processing with Offloading and Energy Harvesting
Modern portable devices can execute increasingly sophisticated AI models on sensed data. The complexity of such processing tasks is data-dependent and has relevant energy cost. This work develops an Age of Information markovian model for a system where multiple battery-operated devices perform data processing and energy harvesting in parallel. Part of their computational burden is offloaded to an edge server which polls devices at given rate. The structural properties of an optimal policy for a single device-server system are derived. They permit to define a new model-free reinforcement learning method specialized for monotone policies, namely Ordered Q-Learning, providing a fast procedure to learn the optimal policy. The method is oblivious to the devices' battery capacities, the cost and the value of data batch processing and to the dynamics of the energy harvesting process. Finally, the polling strategy of the server is optimized by combining this policy improvement technique with stochastic approximation methods. Extensive numerical results provide insight into the system properties and demonstrate that the proposed learning algorithms outperform existing baselines.
- 30 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 83-92
Open AccessLearning Optimal Edge Processing with Offloading and Energy Harvesting
Modern portable devices can execute increasingly sophisticated AI models on sensed data. The complexity of such processing tasks is data-dependent and has relevant energy cost. This work develops an Age of Information markovian model for a system where multiple battery-operated devices perform data processing and energy harvesting in parallel. Part of their computational burden is offloaded to an edge server which polls devices at given rate. The structural properties of an optimal policy for a single device-server system are derived. They permit to define a new model-free reinforcement learning method specialized for monotone policies, namely Ordered Q-Learning, providing a fast procedure to learn the optimal policy. The method is oblivious to the devices' battery capacities, the cost and the value of data batch processing and to the dynamics of the energy harvesting process. Finally, the polling strategy of the server is optimized by combining this policy improvement technique with stochastic approximation methods. Extensive numerical results provide insight into the system properties and demonstrate that the proposed learning algorithms outperform existing baselines.