arXiv subjects

Ali Eslamian

Publications and source records attributed to Ali Eslamian.

8 recordsLinked to original sources

arXiv2026

TabNSM: Neural Sparse Mixer for Tabular Regression

Large-scale, high-dimensional tabular regression remains challenging: tree-based models are robust but lack end-to-end representation learning, while deep models enable flexible feature learning but often incur costly interaction modeling and sensitivity to noisy or redundant features. We propose TabNSM, a scalable regression framework that extends our earlier sparse-attention and mixer architectures. At its core, the Adaptive Sparse Interaction Module (ASIM) integrates foreground feature discovery, sparse local interaction encoding, and Feature-Token Mixing, providing near-linear complexity under fixed sparse configurations. For regression, TabNSM introduces three complementary components: a Multi-Stage Regression Head for progressive prediction refinement; GridLoss, an ordinal-aware soft-binning objective that incorporates target structure into representation learning; and RISE (Reweighted Instance Sampling by Error), a difficulty-aware sampling strategy based on loss-quantile bins. Across nine real-world regression benchmarks, TabNSM delivers strong predictive performance and practical scalability, with particularly consistent gains on high-dimensional and heterogeneous datasets. These results demonstrate that selective interaction modeling, structured regression supervision, and difficulty-aware sampling provide an effective and scalable approach to deep tabular regression.

cs.LG↗

arXiv2026

Predicting When Random Low-Dimensional Reparameterizations Train Neural Networks

Neural networks can often be trained or fine-tuned through random low-dimensional reparameterization, where a small latent vector is mapped into a full parameter update by a frozen random map. This raises a practical question: how large must the latent search space be to reach a low-loss region? We first express the known accessibility transition in an equivalent conic form, centered for compact convex targets at the statistical dimension of the polar cone. Our main theoretical contribution is an orientation-resolved quadratic master formula that predicts the random-slice residual from both the curvature spectrum and the reference-to-solution displacement profile. It yields a self-consistent isotropic-orientation predictor and, in a conservative radius-only specialization, recovers the earlier Gaussian-width quadratic bound. Building on this analysis, we introduce Random Mapping Networks (RaMaN), which instantiate the predicted latent dimension using structured Hadamard or seed-regenerated Gaussian maps. These constructions avoid the O(dP) storage of dense random maps and reduce optimizer-state memory from O(P) to O(d). We also develop matrix-free curvature approximations and sweep-free dimension selection. Across controlled quadratic and neural-curvature experiments, the orientation-resolved predictor closely tracks measured transition locations and outperforms orientation-agnostic approximations when displacement direction matters. End-to-end experiments further show sharp, protocol-dependent training transitions across image and language models.

cs.LG↗

arXiv2025

TabKAN: Advancing Tabular Data Analysis using Kolmogorov-Arnold Network

Tabular data analysis presents unique challenges that arise from heterogeneous feature types, missing values, and complex feature interactions. While traditional machine learning methods like gradient boosting often outperform deep learning, recent advancements in neural architectures offer promising alternatives. In this study, we introduce TabKAN, a novel framework for tabular data modeling based on Kolmogorov-Arnold Networks (KANs). Unlike conventional deep learning models, KANs use learnable activation functions on edges, which improves both interpretability and training efficiency. TabKAN incorporates modular KAN-based architectures designed for tabular analysis and proposes a transfer learning framework for knowledge transfer across domains. Furthermore, we develop a model-specific interpretability approach that reduces reliance on post hoc explanations. Extensive experiments on public datasets show that TabKAN achieves superior performance in supervised learning and significantly outperforms classical and Transformer-based models in binary and multi-class classification. The results demonstrate the potential of KAN-based architectures to bridge the gap between traditional machine learning and deep learning for structured data.

cs.LG↗

arXiv2025

TabNSA: Native Sparse Attention for Efficient Tabular Data Learning

Tabular data poses unique challenges for deep learning due to its heterogeneous feature types, lack of spatial structure, and often limited sample sizes. We propose TabNSA, a novel deep learning framework that integrates Native Sparse Attention (NSA) with a TabMixer backbone to efficiently model tabular data. TabNSA tackles computational and representational challenges by dynamically focusing on relevant feature subsets per instance. The NSA module employs a hierarchical sparse attention mechanism, including token compression, selective preservation, and localized sliding windows, to significantly reduce the quadratic complexity of standard attention operations while addressing feature heterogeneity. Complementing this, the TabMixer backbone captures complex, non-linear dependencies through parallel multilayer perceptron (MLP) branches with independent parameters. These modules are synergistically combined via element-wise summation and mean pooling, enabling TabNSA to model both global context and fine-grained interactions. Extensive experiments across supervised and transfer learning settings show that TabNSA consistently outperforms state-of-the-art deep learning models. Furthermore, by augmenting TabNSA with a fine-tuned large language model (LLM), we enable it to effectively address Few-Shot Learning challenges through language-guided generalization on diverse tabular benchmarks. Code available on: https://github.com/aseslamian/TabNSA

cs.LG↗

arXiv2024

Evaluating Deep Learning Models for Breast Cancer Classification: A Comparative Study

This study evaluates the effectiveness of deep learning models in classifying histopathological images for early and accurate detection of breast cancer. Eight advanced models, including ResNet-50, DenseNet-121, ResNeXt-50, Vision Transformer (ViT), GoogLeNet (Inception v3), EfficientNet, MobileNet, and SqueezeNet, were compared using a dataset of 277,524 image patches. The Vision Transformer (ViT) model, with its attention-based mechanisms, achieved the highest validation accuracy of 94%, outperforming conventional CNNs. The study demonstrates the potential of advanced machine learning methods to enhance precision and efficiency in breast cancer diagnosis in clinical settings.

eess.IV↗

arXiv2023

Enhanced Unscented Kalman Filter-Based SLAM in Dynamic Environments: Euclidean Approach

This paper introduces an innovative approach to Simultaneous Localization and Mapping (SLAM) using the Unscented Kalman Filter (UKF) in a dynamic environment. The UKF is proven to be a robust estimator and demonstrates lower sensitivity to sensor data errors compared to alternative SLAM algorithms. However, conventional algorithms are primarily concerned with stationary landmarks, which might prevent localization in dynamic environments. This paper proposes an Euclidean-based method for handling moving landmarks, calculating and estimating distances between the robot and each moving landmark, and addressing sensor measurement conflicts. The approach is evaluated through simulations in MATLAB and comparing results with the conventional UKF-SLAM algorithm. We also introduce a dataset for filter-based algorithms in dynamic environments, which can be used as a benchmark for evaluating of future algorithms. The outcomes of the proposed algorithm underscore that this simple yet effective approach mitigates the disruptive impact of moving landmarks, as evidenced by a thorough examination involving parameters such as the number of moving and stationary landmarks, waypoints, and computational efficiency. We also evaluated our algorithms in a realistic simulation of a real-world mapping task. This approach allowed us to assess our methods in practical conditions and gain insights for future enhancements. Our algorithm surpassed the performance of all competing methods in the evaluation, showcasing its ability to excel in real-world mapping scenarios.

cs.RO↗

arXiv2023

Improving SCGAN's Similarity Constraint and Learning a Better Disentangled Representation

SCGAN adds a similarity constraint between generated images and conditions as a regularization term on generative adversarial networks. Similarity constraint works as a tutor to instruct the generator network to comprehend the difference of representations based on conditions. We understand how SCGAN works on a deeper level. This understanding makes us realize that the similarity constraint functions like the contrastive loss function. We believe that a model with high understanding and intelligence measures the similarity between images based on their structure and high level features, just like humans do. Two major changes we applied to SCGAN in order to make a modified model are using SSIM to measure similarity between images and applying contrastive loss principles to the similarity constraint. The modified model performs better using FID and FactorVAE metrics. The modified model also has better generalisability compared to other models. Keywords Generative Adversarial Nets, Unsupervised Learning, Disentangled Representation Learning, Contrastive Disentanglement, SSIM

cs.CV↗

arXiv2022

Det-SLAM: A semantic visual SLAM for highly dynamic scenes using Detectron2

According to experts, Simultaneous Localization and Mapping (SLAM) is an intrinsic part of autonomous robotic systems. Several SLAM systems with impressive performance have been invented and used during the last several decades. However, there are still unresolved issues, such as how to deal with moving objects in dynamic situations. Classic SLAM systems depend on the assumption of a static environment, which becomes unworkable in highly dynamic situations. Several methods have been presented to tackle this issue in recent years, but each has its limitations. This research combines the visual SLAM systems ORB-SLAM3 and Detectron2 to present the Det-SLAM system, which employs depth information and semantic segmentation to identify and eradicate dynamic spots to accomplish semantic SLAM for dynamic situations. Evaluation of public TUM datasets indicates that Det-SLAM is more resilient than previous dynamic SLAM systems and can lower the estimated error of camera posture in dynamic indoor scenarios.

cs.RO↗