arXiv subjects

Alireza Olama

Publications and source records attributed to Alireza Olama.

8 recordsLinked to original sources

arXiv2026

Entropy-Regularized Probabilistic Gates for Sparse Model Discovery in Scarce-Data Federated Learning

Federated Learning (FL) is a distributed machine learning (ML) paradigm with collaboration among multiple clients without sharing data. FL is challenging under data heterogeneity and partial client participation. Learning sparse models is useful for communication and computational efficiency in FL, but it is especially difficult in the small-sample high-dimensional regime (d >> N) where optimization can yield parameter configurations that fail to generalize to unseen test data. While magnitude-based pruning doesn't account for uncertainty exploration in the parameter space, a formulation with probabilistic gates and an L0 constraint allows sampling from competing sparse configurations during training. In this work, we study entropy regularization of gate distributions as a mechanism to maintain uncertainty in sparse federated optimization by preventing early commitment to sparse support. We examine its impact under data heterogeneity, client participation heterogeneity, and sparsity. Experiments on synthetic and real-world benchmarks show consistent improvements over federated iterative hard thresholding (Fed-IHT) and pruning after dense federated averaging (FedAvg) training, both in statistical performance on test data and in sparsity recovery accuracy.

cs.LG↗

arXiv2025

Federated Learning With L0 Constraint Via Probabilistic Gates For Sparsity

Federated Learning (FL) is a distributed machine learning setting that requires multiple clients to collaborate on training a model while maintaining data privacy. The unaddressed inherent sparsity in data and models often results in overly dense models and poor generalizability under data and client participation heterogeneity. We propose FL with an L0 constraint on the density of non-zero parameters, achieved through a reparameterization using probabilistic gates and their continuous relaxation: originally proposed for sparsity in centralized machine learning. We show that the objective for L0 constrained stochastic minimization naturally arises from an entropy maximization problem of the stochastic gates and propose an algorithm based on federated stochastic gradient descent for distributed learning. We demonstrate that the target density (rho) of parameters can be achieved in FL, under data and client participation heterogeneity, with minimal loss in statistical performance for linear and non-linear models: Linear regression (LR), Logistic regression (LG), Softmax multi-class classification (MC), Multi-label classification with logistic units (MLC), Convolution Neural Network (CNN) for multi-class classification (MC). We compare the results with a magnitude pruning-based thresholding algorithm for sparsity in FL. Experiments on synthetic data with target density down to rho = 0.05 and publicly available RCV1, MNIST, and EMNIST datasets with target density down to rho = 0.005 demonstrate that our approach is communication-efficient and consistently better in statistical performance.

stat.ML↗

arXiv2025

PruneX: A Hierarchical Communication-Efficient System for Distributed CNN Training with Structured Pruning

Inter-node communication bandwidth increasingly constrains distributed training at scale on multi-node GPU clusters. While compact models are the ultimate deployment target, conventional pruning-aware distributed training systems typically fail to reduce communication overhead because unstructured sparsity cannot be efficiently exploited by highly optimized dense collective primitives. We present PruneX, a distributed data-parallel training system that co-designs pruning algorithms with cluster hierarchy to reduce inter-node bandwidth usage. PruneX introduces the Hierarchical Structured ADMM (H-SADMM) algorithm, which enforces node-level structured sparsity before inter-node synchronization, enabling dynamic buffer compaction that eliminates both zero-valued transmissions and indexing overhead. The system adopts a leader-follower execution model with separated intra-node and inter-node process groups, performing dense collectives on compacted tensors over bandwidth-limited links while confining full synchronization to high-bandwidth intra-node interconnects. Evaluation on ResNet architectures across 64 GPUs demonstrates that PruneX reduces inter-node communication volume by approximately 60% and achieves 6.75x strong scaling speedup, outperforming the dense baseline (5.81x) and Top-K gradient compression (3.71x) on the Puhti supercomputer at CSC - IT Center for Science (Finland).

cs.DC↗

arXiv2025

Adaptive Factor Graph-Based Tightly Coupled GNSS/IMU Fusion for Robust Positionin

Reliable positioning in GNSS-challenged environments remains a critical challenge for navigation systems. Tightly coupled GNSS/IMU fusion improves robustness but remains vulnerable to non-Gaussian noise and outliers. We present a robust and adaptive factor graph-based fusion framework that directly integrates GNSS pseudorange measurements with IMU preintegration factors and incorporates the Barron loss, a general robust loss function that unifies several m-estimators through a single tunable parameter. By adaptively down weighting unreliable GNSS measurements, our approach improves resilience positioning. The method is implemented in an extended GTSAM framework and evaluated on the UrbanNav dataset. The proposed solution reduces positioning errors by up to 41% relative to standard FGO, and achieves even larger improvements over extended Kalman filter (EKF) baselines in urban canyon environments. These results highlight the benefits of Barron loss in enhancing the resilience of GNSS/IMU-based navigation in urban and signal-compromised environments.

cs.RO↗

arXiv2024

A GPU-Accelerated Bi-linear ADMM Algorithm for Distributed Sparse Machine Learning

This paper introduces the Bi-linear consensus Alternating Direction Method of Multipliers (Bi-cADMM), aimed at solving large-scale regularized Sparse Machine Learning (SML) problems defined over a network of computational nodes. Mathematically, these are stated as minimization problems with convex local loss functions over a global decision vector, subject to an explicit $\ell_0$ norm constraint to enforce the desired sparsity. The considered SML problem generalizes different sparse regression and classification models, such as sparse linear and logistic regression, sparse softmax regression, and sparse support vector machines. Bi-cADMM leverages a bi-linear consensus reformulation of the original non-convex SML problem and a hierarchical decomposition strategy that divides the problem into smaller sub-problems amenable to parallel computing. In Bi-cADMM, this decomposition strategy is based on a two-phase approach. Initially, it performs a sample decomposition of the data and distributes local datasets across computational nodes. Subsequently, a delayed feature decomposition of the data is conducted on Graphics Processing Units (GPUs) available to each node. This methodology allows Bi-cADMM to undertake computationally intensive data-centric computations on GPUs, while CPUs handle more cost-effective computations. The proposed algorithm is implemented within an open-source Python package called Parallel Sparse Fitting Toolbox (PsFiT), which is publicly available. Finally, computational experiments demonstrate the efficiency and scalability of our algorithm through numerical benchmarks across various SML problems featuring distributed datasets.

cs.LG↗

arXiv2022

Sparse Convex Optimization Toolkit: A Mixed-Integer Framework

This paper proposes an open-source distributed solver for solving Sparse Convex Optimization (SCO) problems over computational networks. Motivated by past algorithmic advances in mixed-integer optimization, the Sparse Convex Optimization Toolkit (SCOT) adopts a mixed-integer approach to find exact solutions to SCO problems. In particular, SCOT brings together various techniques to transform the original SCO problem into an equivalent convex Mixed-Integer Nonlinear Programming (MINLP) problem that can benefit from high-performance and parallel computing platforms. To solve the equivalent mixed-integer problem, we present the Distributed Hybrid Outer Approximation (DiHOA) algorithm that builds upon the LP/NLP based branch-and-bound and is tailored for this specific problem structure. The DiHOA algorithm combines the so-called single- and multi-tree outer approximation, naturally integrates a decentralized algorithm for distributed convex nonlinear subproblems, and utilizes enhancement techniques such as quadratic cuts. Finally, we present detailed computational experiments that show the benefit of our solver through numerical benchmarks on 140 SCO problems with distributed datasets. To show the overall efficiency of SCOT we also provide performance profiles comparing SCOT to other state-of-the-art MINLP solvers.

math.OC↗

arXiv2022

Distributed Primal Outer Approximation Algorithm for Sparse Convex Programming with Separable Structures

This paper presents the Distributed Primal Outer Approximation (DiPOA) algorithm for solving Sparse Convex Programming (SCP) problems with separable structures, efficiently, and in a decentralized manner. The DiPOA algorithm development consists of embedding the recently proposed Relaxed Hybrid Alternating Direction Method of Multipliers (RH-ADMM) algorithm into the Outer Approximation (OA) algorithm. We also propose two main improvements to control the quality and the number of cutting planes that approximate nonlinear functions. In particular, the RH-ADMM algorithm acts as a distributed numerical engine inside the DiPOA algorithm. DiPOA takes advantage of the multi-core architecture of modern processors to speed up optimization algorithms. The proposed distributed algorithm makes practical the solution of SCP in learning and control problems from the application side. This paper concludes with a performance analysis of DiPOA for the distributed sparse logistic regression and quadratically constrained optimization problems. Finally, the paper concludes with a numerical comparison with state-of-the-art optimization solvers.

math.OC↗

arXiv2018

Stable MPC Design for Hybrid Mixed Logical Dynamical Systems: $l_{\infty}$-based Lyapunov Approach

There are two main challenges in control of hybrid systems which are to guarantee the closed-loop stability and reduce computational complexity. In this paper, we propose the exponential stability conditions of hybrid systems which are described in the Mixed Logical Dynamical (MLD) form in closed-loop with Model Predictive Control (MPC). To do this, it is proposed to use the decreasing condition of infinity norm based Lyapunov function instead of imposing the terminal equality constraint in the MPC formulation of MLD system. The exponential stability conditions have a better performance from both implementation and computational points of view. In addition, the exponential stability conditions of the equilibrium point of the MLD system do not depend on the prediction horizon of MPC problem which is the main advantage of the proposed method. On the other hand, by using the decreasing condition of the Lyapunov function in the MPC setup, the suboptimal version of the control signal with reduced complexity is obtained. In order to show the capabilities of the proposed method, the stabilization problem of the car suspension system is studied.

eess.SY↗