Ali Sekmen

Sakib Abrar, Ali Sekmen, Manar D. Samad

Deep learning methods in the literature are commonly benchmarked on image data sets, which may not be suitable or effective baselines for non-image tabular data. In this paper, we take a data-centric view to perform one of the first studies on deep embedding clustering of tabular data. Eight clustering and state-of-the-art embedding clustering methods proposed for image data sets are tested on seven tabular data sets. Our results reveal that a traditional clustering method ranks second out of eight methods and is superior to most deep embedding clustering baselines. Our observation aligns with the literature that conventional machine learning of tabular data is still a robust approach against deep learning. Therefore, state-of-the-art embedding clustering methods should consider data-centric customization of learning architectures to become competitive baselines for tabular data.

Tim Wallace, Ali Sekmen

Kaczmarz's alternating projection method has been widely used for solving a consistent (mostly over-determined) linear system of equations Ax=b. Because of its simple iterative nature with light computation, this method was successfully applied in computerized tomography. Since tomography generates a matrix A with highly coherent rows, randomized Kaczmarz algorithm is expected to provide faster convergence as it picks a row for each iteration at random, based on a certain probability distribution. It was recently shown that picking a row at random, proportional with its norm, makes the iteration converge exponentially in expectation with a decay constant that depends on the scaled condition number of A and not the number of equations. Since Kaczmarz's method is a subspace projection method, the convergence rate for simple Kaczmarz algorithm was developed in terms of subspace angles. This paper provides analyses of simple and randomized Kaczmarz algorithms and explain the link between them. It also propose new versions of randomization that may speed up convergence.

Zeinab Elkhatib, Ali Sekmen, Kamrul Hasan

With the rapid advancements in machine learning, models have become increasingly capable of learning and making predictions in various industries. However, deploying these models in critical infrastructures presents a major challenge, as concerns about data privacy prevent unrestricted data sharing. Homomorphic encryption (HE) offers a solution by enabling computations on encrypted data, but it remains incompatible with machine learning models like convolutional neural networks (CNNs), due to their reliance on non-linear activation functions. To bridge this gap, this work proposes an optimized framework that replaces standard non-linear functions with homomorphically compatible approximations, ensuring secure computations while minimizing computational overhead. The proposed approach restructures the CNN architecture and introduces an efficient activation function approximation method to mitigate the performance trade-offs introduced by encryption. Experiments on CIFAR-10 achieve 94.4% accuracy with 2.42 s per single encrypted sample and 24,000 s per 10,000 encrypted samples, using a degree-4 polynomial and Softplus activation under CKKS, balancing accuracy and privacy.

Akram Aldroubi, Keaton Hamm, Ahmet Bugra Koku et al.

A general framework for solving the subspace clustering problem using the CUR decomposition is presented. The CUR decomposition provides a natural way to construct similarity matrices for data that come from a union of unknown subspaces $\mathscr{U}=\underset{i=1}{\overset{M}\bigcup}S_i$. The similarity matrices thus constructed give the exact clustering in the noise-free case. Additionally, this decomposition gives rise to many distinct similarity matrices from a given set of data, which allow enough flexibility to perform accurate clustering of noisy data. We also show that two known methods for subspace clustering can be derived from the CUR decomposition. An algorithm based on the theoretical construction of similarity matrices is presented, and experiments on synthetic and real data are presented to test the method. Additionally, an adaptation of our CUR based similarity matrices is utilized to provide a heuristic algorithm for subspace clustering; this algorithm yields the best overall performance to date for clustering the Hopkins155 motion segmentation dataset.