György Dán
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
György Dán has a total of 122 co-authors. Here are the top co-authors by number of shared publications:
- Sladana Josilo
Division of Network and Systems Engineering, School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology, Stockholm, Sweden
9 shared publications - Ezzeldin Shereen
School of Electrical Engineering, Computer Science, KTH Royal Institute of Technology, Stockholm, Sweden
7 shared publications - Henrik Sandberg
KTH Royal Institute of Technology
6 shared publications - Mohammad Reza Hesamzadeh
KTH Royal Institute of Technology
6 shared publications - Feridun Tütüncüoğlu
Division of Network and Systems Engineering, School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology, Stockholm, Sweden
5 shared publications - Emil Eriksson
School of Electrical Engineering, KTH Royal Institute of Technology, Stockholm, Sweden
4 shared publications
Latest Publications
- Open Access
- 13 November 2026
- Published by Elsevier in Energy Strategy Reviews
- Vol. 68
Interpretable system-level forecasting of Nordic day-ahead electricity prices using PCA-augmented selection-shrinkage
- 13 November 2026
- Published by Elsevier in Energy Strategy Reviews
- Vol. 68
Open AccessInterpretable system-level forecasting of Nordic day-ahead electricity prices using PCA-augmented selection-shrinkage
- Open Access
- 23 September 2026
- Published by Elsevier in Electric Power Systems Research
- Vol. 265
A tax-subsidy scheme for efficient investment in renewable generation capacity
- 23 September 2026
- Published by Elsevier in Electric Power Systems Research
- Vol. 265
Open AccessA tax-subsidy scheme for efficient investment in renewable generation capacity
- 1 April 2026
- Published by Springer Nature
- p. 246-265
Mitigating Traffic Analysis Attacks While Maintaining On-Path Network Observability
- 1 April 2026
- Published by Springer Nature
- p. 246-265
Mitigating Traffic Analysis Attacks While Maintaining On-Path Network Observability
- Open Access
- 23 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 34, 4403-4415
Revenue Optimal Orchestration of ML-Based Services With Dependencies Under Delay and Quality Constraints in Beyond 5G RAN
Effective service deployment and orchestration will be essential to accommodate user workloads with diverse requirements in cloud-native beyond 5G Radio Access Networks (RAN). Orchestration will have to take into account individual service quality requirements, latency constraints, and dependencies, while leveraging unique characteristics of dominant workloads, such as machine learning (ML) models. In this work, we address the orchestration of ML-based services, considering users that request application services that rely on network services, such as localization, positioning, etc. Each service is composed of functions, at potentially different quality levels. The objective is to maximize the network operator’s revenue by determining service deployment, quality selection and computational resource allocation. The resulting problem is a mixed-integer non-convex problem, which we show is NP-hard. We provide sufficient conditions for the problem to be submodular, and for the general case we propose JADES, which relies on linear relaxation and convexification to decompose the problem into two subproblems, which are solved iteratively until convergence, followed by dependent randomized rounding. Our evaluation based on synthetic workloads shows that JADES outperforms baselines in terms of operator revenue and computational efficiency.
- 23 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 34, 4403-4415
Open AccessRevenue Optimal Orchestration of ML-Based Services With Dependencies Under Delay and Quality Constraints in Beyond 5G RAN
Effective service deployment and orchestration will be essential to accommodate user workloads with diverse requirements in cloud-native beyond 5G Radio Access Networks (RAN). Orchestration will have to take into account individual service quality requirements, latency constraints, and dependencies, while leveraging unique characteristics of dominant workloads, such as machine learning (ML) models. In this work, we address the orchestration of ML-based services, considering users that request application services that rely on network services, such as localization, positioning, etc. Each service is composed of functions, at potentially different quality levels. The objective is to maximize the network operator’s revenue by determining service deployment, quality selection and computational resource allocation. The resulting problem is a mixed-integer non-convex problem, which we show is NP-hard. We provide sufficient conditions for the problem to be submodular, and for the general case we propose JADES, which relies on linear relaxation and convexification to decompose the problem into two subproblems, which are solved iteratively until convergence, followed by dependent randomized rounding. Our evaluation based on synthetic workloads shows that JADES outperforms baselines in terms of operator revenue and computational efficiency.
- Open Access
- 2 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 34, 4333-4344
RAPTOR: Rate-Adaptive Pricing and Optimal Resource Allocation in Serverless Edge Computing
Edge computing (EC) is emerging as a key enabler for latency-sensitive applications such as Augmented Reality (AR), autonomous driving, and industrial IoT, by bringing computational resources closer to Wireless Devices (WDs). However, the limited computational capacity inherent to EC presents challenges in resource allocation and in designing pricing mechanisms that provide the right incentives and are aligned with the user-perceived service quality. This paper addresses these challenges by formulating a Stackelberg game that models WDs’ valuation of EC services based on their offloading rates and the service quality they receive. We prove the existence of Stackelberg equilibria and we propose a tractable approximation technique based on log-barrier functions for computing approximate equilibria. Furthermore, to overcome computational issues, we build on the concept of a Differential Stackelberg Equilibrium (DSE) and we propose Stackelberg Gradient Play (SGP), an implicit gradient-based algorithm that ensures convergence to DSE while maintaining efficiency. Extensive simulations show that our approach significantly outperforms existing methods, achieving up to 70% higher revenue for the edge operator while reducing computational overhead substantially. These results underscore the viability of our framework for use in EC systems that require fast, adaptive, and service-aware joint resource management and pricing.
- 2 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 34, 4333-4344
Open AccessRAPTOR: Rate-Adaptive Pricing and Optimal Resource Allocation in Serverless Edge Computing
Edge computing (EC) is emerging as a key enabler for latency-sensitive applications such as Augmented Reality (AR), autonomous driving, and industrial IoT, by bringing computational resources closer to Wireless Devices (WDs). However, the limited computational capacity inherent to EC presents challenges in resource allocation and in designing pricing mechanisms that provide the right incentives and are aligned with the user-perceived service quality. This paper addresses these challenges by formulating a Stackelberg game that models WDs’ valuation of EC services based on their offloading rates and the service quality they receive. We prove the existence of Stackelberg equilibria and we propose a tractable approximation technique based on log-barrier functions for computing approximate equilibria. Furthermore, to overcome computational issues, we build on the concept of a Differential Stackelberg Equilibrium (DSE) and we propose Stackelberg Gradient Play (SGP), an implicit gradient-based algorithm that ensures convergence to DSE while maintaining efficiency. Extensive simulations show that our approach significantly outperforms existing methods, achieving up to 70% higher revenue for the edge operator while reducing computational overhead substantially. These results underscore the viability of our framework for use in EC systems that require fast, adaptive, and service-aware joint resource management and pricing.
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Energy Markets, Policy and Regulation
- Vol. 4 (3) , 1-14
Contract Design for Facilitating Risk Management in Local Electricity Markets
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Energy Markets, Policy and Regulation
- Vol. 4 (3) , 1-14
Open AccessContract Design for Facilitating Risk Management in Local Electricity Markets
- Open Access
Driver Identification and PCA Augmented Selection Shrinkage Framework for Nordic System Price Forecasting
The System Price (SP) of the Nordic electricity market serves as a key reference for financial hedge contracts such as Electricity Price Area Differentials (EPADs) and other risk management instruments. Therefore, the identification of drivers and the accurate forecasting of SP are essential for market participants to design effective hedging strategies. This paper develops a systematic framework that combines interpretable drivers analysis with robust forecasting methods. It proposes an interpretable feature engineering algorithm to identify the main drivers of the Nordic SP based on a novel combination of K-means clustering, Multiple Seasonal-Trend Decomposition (MSTD), and Seasonal Autoregressive Integrated Moving Average (SARIMA) model. Then, it applies principal component analysis (PCA) to the identified data matrix, which is adapted to the downstream task of price forecasting to mitigate the issue of imperfect multicollinearity in the data. Finally, we propose a multi-forecast selection-shrinkage algorithm for Nordic SP forecasting, which selects a subset of complementary forecast models based on their bias-variance tradeoff at the ensemble level and then computes the optimal weights for the retained forecast models to minimize the error variance of the combined forecast. Using historical data from the Nordic electricity market, we demonstrate that the proposed approach outperforms individual input models uniformly, robustly, and significantly, while maintaining a comparable computational cost. Notably, our systematic framework produces superior results using simple input models, outperforming the state-of-the-art Temporal Fusion Transformer (TFT). Furthermore, we show that our approach also exceeds the performance of several well-established practical forecast combination methods.
Open AccessDriver Identification and PCA Augmented Selection Shrinkage Framework for Nordic System Price Forecasting
The System Price (SP) of the Nordic electricity market serves as a key reference for financial hedge contracts such as Electricity Price Area Differentials (EPADs) and other risk management instruments. Therefore, the identification of drivers and the accurate forecasting of SP are essential for market participants to design effective hedging strategies. This paper develops a systematic framework that combines interpretable drivers analysis with robust forecasting methods. It proposes an interpretable feature engineering algorithm to identify the main drivers of the Nordic SP based on a novel combination of K-means clustering, Multiple Seasonal-Trend Decomposition (MSTD), and Seasonal Autoregressive Integrated Moving Average (SARIMA) model. Then, it applies principal component analysis (PCA) to the identified data matrix, which is adapted to the downstream task of price forecasting to mitigate the issue of imperfect multicollinearity in the data. Finally, we propose a multi-forecast selection-shrinkage algorithm for Nordic SP forecasting, which selects a subset of complementary forecast models based on their bias-variance tradeoff at the ensemble level and then computes the optimal weights for the retained forecast models to minimize the error variance of the combined forecast. Using historical data from the Nordic electricity market, we demonstrate that the proposed approach outperforms individual input models uniformly, robustly, and significantly, while maintaining a comparable computational cost. Notably, our systematic framework produces superior results using simple input models, outperforming the state-of-the-art Temporal Fusion Transformer (TFT). Furthermore, we show that our approach also exceeds the performance of several well-established practical forecast combination methods.
- Open Access
- 24 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 22 (6) , 6913-6928
Dynamic Alert Prioritization for Real-Time Situational Awareness: A Hidden Markov Model Framework With Active Learning
Real-time cyber situational awareness (SA) is crucial for effective and timely incident response. However, maintaining SA requires substantial human effort; security analysts must analyze large volumes of alerts, many of which are false positives triggered by anomaly-based intrusion detection systems (IDSs). Efficiently prioritizing these alerts is vital to enable analysts to focus on real threats without delay. In this paper, we present two key contributions designed to improve real-time SA. First, we propose modeling dynamic alert prioritization as an active learning problem in a hidden Markov model (HMM) with the objective to minimize the mean squared error (MSE) of the belief. We propose to use the uncertainty of the belief as a proxy for the MSE of the belief, and we develop two computationally tractable policies for choosing alerts to investigate. Second, we propose and evaluate a state space and an exploit space reduction method to reduce the computational complexity of the belief update. We use simulations on synthetic and real dependency graphs to evaluate the proposed policies. Our results show that the proposed investigation policies reduce the MSE of the belief by up to 50% compared to baseline policies, and they are robust to high false alert rates and to investigation errors. Our results also show that state space reduction can reduce the computation time by 85% without a significant increase in the belief MSE.
- 24 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 22 (6) , 6913-6928
Open AccessDynamic Alert Prioritization for Real-Time Situational Awareness: A Hidden Markov Model Framework With Active Learning
Real-time cyber situational awareness (SA) is crucial for effective and timely incident response. However, maintaining SA requires substantial human effort; security analysts must analyze large volumes of alerts, many of which are false positives triggered by anomaly-based intrusion detection systems (IDSs). Efficiently prioritizing these alerts is vital to enable analysts to focus on real threats without delay. In this paper, we present two key contributions designed to improve real-time SA. First, we propose modeling dynamic alert prioritization as an active learning problem in a hidden Markov model (HMM) with the objective to minimize the mean squared error (MSE) of the belief. We propose to use the uncertainty of the belief as a proxy for the MSE of the belief, and we develop two computationally tractable policies for choosing alerts to investigate. Second, we propose and evaluate a state space and an exploit space reduction method to reduce the computational complexity of the belief update. We use simulations on synthetic and real dependency graphs to evaluate the proposed policies. Our results show that the proposed investigation policies reduce the MSE of the belief by up to 50% compared to baseline policies, and they are robust to high false alert rates and to investigation errors. Our results also show that state space reduction can reduce the computation time by 85% without a significant increase in the belief MSE.
- Open Access
Quickest Detection of Adversarial Attacks Against Correlated Equilibria
We consider correlated equilibria in strategic games in an adversarial environment, where an adversary can compromise the public signal used by the players for choosing their strategies, while players aim at detecting a potential attack as soon as possible to avoid loss of utility. We model the interaction between the adversary and the players as a zero-sum game and we derive the maxmin strategies for both the defender and the attacker using the framework of quickest change detection. We define a class of adversarial strategies that achieve the optimal trade-off between attack impact and attack detectability and show that a generalized CUSUM scheme is asymptotically optimal for the detection of the attacks. Our numerical results on the Sioux-Falls benchmark traffic routing game show that the proposed detection scheme can effectively limit the utility loss by a potential adversary.
Open AccessQuickest Detection of Adversarial Attacks Against Correlated Equilibria
We consider correlated equilibria in strategic games in an adversarial environment, where an adversary can compromise the public signal used by the players for choosing their strategies, while players aim at detecting a potential attack as soon as possible to avoid loss of utility. We model the interaction between the adversary and the players as a zero-sum game and we derive the maxmin strategies for both the defender and the attacker using the framework of quickest change detection. We define a class of adversarial strategies that achieve the optimal trade-off between attack impact and attack detectability and show that a generalized CUSUM scheme is asymptotically optimal for the detection of the attacks. Our numerical results on the Sioux-Falls benchmark traffic routing game show that the proposed detection scheme can effectively limit the utility loss by a potential adversary.
- Open Access
SpaNN: Detecting Multiple Adversarial Patches on CNNs by Spanning Saliency Thresholds
State-of-the-art convolutional neural network models for object detection and image classification are vulnerable to physically realizable adversarial perturbations, such as patch attacks. Existing defenses have focused, implicitly or explicitly, on single-patch attacks, leaving their sensitivity to the number of patches as an open question or rendering them computationally infeasible or inefficient against attacks consisting of multiple patches in the worst cases. In this work, we propose SpaNN, an attack detector whose computational complexity is independent of the expected number of adversarial patches. The key novelty of the proposed detector is that it builds an ensemble of binarized feature maps by applying a set of saliency thresholds to the neural activations of the first convolutional layer of the victim model. It then performs clustering on the ensemble and uses the cluster features as the input to a classifier for attack detection. Contrary to existing detectors, SpaNN does not rely on a fixed saliency threshold for identifying adversarial regions, which makes it robust against white box adversarial attacks. We evaluate SpaNN on four widely used data sets for object detection and classification, and our results show that SpaNN outperforms state-of-the-art defenses by up to 11 and 27 percentage points in the case of object detection and the case of image classification, respectively. Our code is available at https://github.com/gerkbyrd/SpaNN.
Open AccessSpaNN: Detecting Multiple Adversarial Patches on CNNs by Spanning Saliency Thresholds
State-of-the-art convolutional neural network models for object detection and image classification are vulnerable to physically realizable adversarial perturbations, such as patch attacks. Existing defenses have focused, implicitly or explicitly, on single-patch attacks, leaving their sensitivity to the number of patches as an open question or rendering them computationally infeasible or inefficient against attacks consisting of multiple patches in the worst cases. In this work, we propose SpaNN, an attack detector whose computational complexity is independent of the expected number of adversarial patches. The key novelty of the proposed detector is that it builds an ensemble of binarized feature maps by applying a set of saliency thresholds to the neural activations of the first convolutional layer of the victim model. It then performs clustering on the ensemble and uses the cluster features as the input to a classifier for attack detection. Contrary to existing detectors, SpaNN does not rely on a fixed saliency threshold for identifying adversarial regions, which makes it robust against white box adversarial attacks. We evaluate SpaNN on four widely used data sets for object detection and classification, and our results show that SpaNN outperforms state-of-the-art defenses by up to 11 and 27 percentage points in the case of object detection and the case of image classification, respectively. Our code is available at https://github.com/gerkbyrd/SpaNN.