Aleksandar Armacki

21 papers A* 1A 1Journal 14Unranked 5

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	Distributed Gradient Clustering: Convergence and the Effect of Initialization. CoRR Aleksandar Armacki, Himkant Sharma, Dragana Bajovic, Dusan Jakovetic, Mrityunjoy Chakraborty, Soummya Kar
2026	J	`jnl`	Large Deviation Upper Bounds and Improved MSE Rates of Nonlinear SGD: Heavy-Tailed Noise and Power of Symmetry. SIAM J. Optim. Aleksandar Armacki, Shuhua Yu, Dragana Bajovic, Dusan Jakovetic, Soummya Kar
2026	J	`jnl`	Tight Long-Term Tail Decay of (Clipped) SGD in Non-Convex Optimization. CoRR Aleksandar Armacki, Dragana Bajovic, Dusan Jakovetic, Soummya Kar, Ali H. Sayed
2025	J	`jnl`	Distributed Center-Based Clustering: A Unified Framework. IEEE Trans. Signal Process. Aleksandar Armacki, Dragana Bajovic, Dusan Jakovetic, Soummya Kar
2025	A	`conf`	High-probability Convergence Bounds for Online Nonlinear Stochastic Gradient Descent under Heavy-tailed Noise. AISTATS Aleksandar Armacki, Shuhua Yu, Pranay Sharma, Gauri Joshi, Dragana Bajovic, Dusan Jakovetic, Soummya Kar
2025	J	`jnl`	Improved High-probability Convergence Guarantees of Decentralized SGD. CoRR Aleksandar Armacki, Ali H. Sayed
2025	J	`jnl`	Optimal High-probability Convergence of Nonlinear SGD under Heavy-tailed Noise via Symmetrization. CoRR Aleksandar Armacki, Dragana Bajovic, Dusan Jakovetic, Soummya Kar
2025	—	`conf`	Toward Understanding the Improved Robustness to Initialization in Distributed Clustering. EUSIPCO Aleksandar Armacki, Soummya Kar
2024	J	`jnl`	A One-Shot Framework for Distributed Clustered Learning in Heterogeneous Environments. IEEE Trans. Signal Process. Aleksandar Armacki, Dragana Bajovic, Dusan Jakovetic, Soummya Kar
2024	J	`jnl`	A Unified Framework for Gradient-based Clustering of Distributed Data. CoRR Aleksandar Armacki, Dragana Bajovic, Dusan Jakovetic, Soummya Kar
2024	—	`conf`	Distributed Gradient Clustering: Convergence and the Effect of Initialization. IEEECONF Aleksandar Armacki, Himkant Sharma, Dragana Bajovic, Dusan Jakovetic, Mrityunjoy Chakraborty, Soummya Kar
2024	J	`jnl`	Large Deviations and Improved Mean-squared Error Rates of Nonlinear SGD: Heavy-tailed Noise and Power of Symmetry. CoRR Aleksandar Armacki, Shuhua Yu, Dragana Bajovic, Dusan Jakovetic, Soummya Kar
2024	J	`jnl`	Nonlinear Stochastic Gradient Descent and Heavy-tailed Noise: A Unified Framework and High-probability Guarantees. CoRR Aleksandar Armacki, Shuhua Yu, Pranay Sharma, Gauri Joshi, Dragana Bajovic, Dusan Jakovetic, Soummya Kar
2023	—	`conf`	Communication Efficient Model-Aware Federated Learning for Visual Crowd Counting and Density Estimation in Smart Cities. EUSIPCO Aleksandar Armacki, Nemanja Milosevic, Dragana Bajovic, Soummya Kar, Dusan Jakovetic, Arian Bakhtiarnia, Lukas Esterle, Adrian Muscat, Thomas Festi
2023	J	`jnl`	High-probability Convergence Bounds for Nonlinear Stochastic Gradient Descent Under Heavy-tailed Noise. CoRR Aleksandar Armacki, Pranay Sharma, Gauri Joshi, Dragana Bajovic, Dusan Jakovetic, Soummya Kar
2022	A*	`conf`	Gradient Based Clustering. ICML Aleksandar Armacki, Dragana Bajovic, Dusan Jakovetic, Soummya Kar
2022	J	`jnl`	Gradient Based Clustering. CoRR Aleksandar Armacki, Dragana Bajovic, Dusan Jakovetic, Soummya Kar
2022	J	`jnl`	One-Shot Federated Learning for Model Clustering and Learning in Heterogeneous Environments. CoRR Aleksandar Armacki, Dragana Bajovic, Dusan Jakovetic, Soummya Kar
2022	—	`conf`	Personalized Federated Learning via Convex Clustering. ISC2 Aleksandar Armacki, Dragana Bajovic, Dusan Jakovetic, Soummya Kar
2022	J	`jnl`	Personalized Federated Learning via Convex Clustering. CoRR Aleksandar Armacki, Dragana Bajovic, Dusan Jakovetic, Soummya Kar
2019	—	`conf`	Distributed Trust-Region Method With First Order Models. EUROCON Aleksandar Armacki, Dusan Jakovetic, Natasa Krejic, Natasa Krklec Jerinkic