arXiv subjects

Alhassan S. Yasin

Publications and source records attributed to Alhassan S. Yasin.

11 recordsLinked to original sources

arXiv2026

System-Self as a Data Structure: An Architectural Approach to Bounded Adaptation

Safety critical autonomous systems often adapt by adjusting controller parameters while keeping the underlying architecture fixed. This strategy breaks down when shifts in sensing, resource availability, or component health invalidate the original structural assumptions. This work introduces a method in which system maintain an explicit, graph-based representation of their architecture and reason over it during operation. The system is modeled as a directed graph of physical, functional, and model based modules, with edges capturing information and control dependencies. Adaptation is posed as a joint optimization over architectural configurations and module parameters, subject to operational constraints using a Monitor-Analyze-Plan-Execute loop-based finite state machine. Performance degradation is isolated via residual decomposition and dependency weighted influence propagation, and candidate adaptations are filtered using a stability aware mechanism. The approach is demonstrated on a differential drive robot under sensor drift and actuator faults. A fixed architecture accumulates tracking errors of up to 24 m and 13 m, respectively, whereas architecture aware adaptation reduces error under 1.5 m in each case, by selecting fault appropriate configurations. These results show the value of reasoning over system structure, while preserving stability, rather than relying solely on parameter tuning.

eess.SY↗

arXiv2026

Improving Artifact Robustness for CT Deep Learning Models Without Labeled Artifact Images via Domain Adaptation

If a CT scanner introduces a new artifact not present in the training labels, the model may misclassify the images. Although modern CT scanners include design features which mitigate these artifacts, unanticipated or difficult-to-mitigate artifacts can still appear in practice. The direct solution of labeling images from this new distribution can be costly. As a more accessible alternative, this study evaluates domain adaptation as an approach for training models that maintain classification performance despite new artifacts, even without corresponding labels. We simulate ring artifacts from detector gain error in sinogram space and evaluate domain adversarial neural networks (DANN) against baseline and augmentation-based approaches on the OrganAMNIST abdominal CT dataset. We simulate the absence of labels from an unseen distribution via masking in the loss function and selectively detaching unlabeled instances from the computational graph. Our results demonstrate that baseline models trained only on clean images fail to generalize to images with ring artifacts, and traditional augmentation with other distortion types provides no improvement on unseen artifact domains. In contrast, the DANN approach improves classification accuracy on ring artifact images using only unlabeled artifact data during training, demonstrating the viability of domain adaptation for artifact robustness. The domain-adapted model achieved a classification accuracy of 77.4% on ring artifact test data, 38.7% higher than a baseline model only trained on images with no artifact. These findings provide empirical evidence that domain adaptation can effectively address distribution shift in medical imaging without requiring expensive expert labeling of new artifact distributions, suggesting promise for deployment in clinical settings where novel artifacts may emerge.

cs.CV↗

arXiv2026

Transfer Learning from One Cancer to Another via Deep Learning Domain Adaptation

Supervised deep learning models often achieve excellent performance within their training distribution but struggle to generalize beyond it. In cancer histopathology, for example, a convolutional neural network (CNN) may classify cancer severity accurately for cancer types represented in its training data, yet fail on related but unseen types. Although adenocarcinomas from different organs share morphological features that might support limited cross-domain generalization, addressing domain shift directly is necessary for robust performance. Domain adaptation offers a way to transfer knowledge from labeled data in one cancer type to unlabeled data in another, helping mitigate the scarcity of annotated medical images. This work evaluates cross-domain classification performance among lung, colon, breast, and kidney adenocarcinomas. A ResNet50 trained on any single adenocarcinoma achieves over 98% accuracy on its own domain but shows minimal generalization to others. Ensembling multiple supervised models does not resolve this limitation. In contrast, converting the ResNet50 into a domain adversarial neural network (DANN) substantially improves performance on unlabeled target domains. A DANN trained on labeled breast and colon data and adapted to unlabeled lung data reaches 95.56% accuracy. We also examine the impact of stain normalization on domain adaptation. Its effects vary by target domain: for lung, accuracy drops from 95.56% to 66.60%, while for breast and colon targets, stain normalization boosts accuracy from 49.22% to 81.29% and from 78.48% to 83.36%, respectively. Finally, using Integrated Gradients reveals that DANNs consistently attribute importance to biologically meaningful regions such as densely packed nuclei, indicating that the model learns clinically relevant features and can apply them to unlabeled cancer types.

cs.CV↗

arXiv2026

Proximal Policy Optimization with Evolutionary Mutations

Proximal Policy Optimization (PPO) is a widely used reinforcement learning algorithm known for its stability and sample efficiency, but it often suffers from premature convergence due to limited exploration. In this paper, we propose POEM (Proximal Policy Optimization with Evolutionary Mutations), a novel modification to PPO that introduces an adaptive exploration mechanism inspired by evolutionary algorithms. POEM enhances policy diversity by monitoring the Kullback-Leibler (KL) divergence between the current policy and a moving average of previous policies. When policy changes become minimal, indicating stagnation, POEM triggers an adaptive mutation of policy parameters to promote exploration. We evaluate POEM on four OpenAI Gym environments: CarRacing, MountainCar, BipedalWalker, and LunarLander. Through extensive fine-tuning using Bayesian optimization techniques and statistical testing using Welch's t-test, we find that POEM significantly outperforms PPO on three of the four tasks (BipedalWalker: t=-2.0642, p=0.0495; CarRacing: t=-6.3987, p=0.0002; MountainCar: t=-6.2431, p<0.0001), while performance on LunarLander is not statistically significant (t=-1.8707, p=0.0778). Our results highlight the potential of integrating evolutionary principles into policy gradient methods to overcome exploration-exploitation tradeoffs.

cs.NE↗

arXiv2024

QR Sort: A Novel Non-Comparative Sorting Algorithm

In this paper, we introduce and prove QR Sort, a novel non-comparative integer sorting algorithm. This algorithm uses principles derived from the Quotient-Remainder Theorem and Counting Sort subroutines to sort input sequences stably. QR Sort exhibits the general time and space complexity $\mathcal{O}(n+d+\frac{m}{d})$, where $n$ denotes the input sequence length, $d$ denotes a predetermined positive integer, and $m$ denotes the range of input sequence values plus 1. Setting $d = \sqrt{m}$ minimizes time and space to $\mathcal{O}(n + \sqrt{m})$, resulting in linear time and space $\mathcal{O}(n)$ when $m \leq \mathcal{O}(n^2)$. We provide implementation optimizations for minimizing the time and space complexity, runtime, and number of computations expended by QR Sort, showcasing its adaptability. Our results reveal that QR Sort frequently outperforms established algorithms and serves as a reliable sorting algorithm for input sequences that exhibit large $m$ relative to $n$.

cs.DS↗

arXiv2024

Reinforcement Learning Framework for Quantitative Trading

The inherent volatility and dynamic fluctuations within the financial stock market underscore the necessity for investors to employ a comprehensive and reliable approach that integrates risk management strategies, market trends, and the movement trends of individual securities. By evaluating specific data, investors can make more informed decisions. However, the current body of literature lacks substantial evidence supporting the practical efficacy of reinforcement learning (RL) agents, as many models have only demonstrated success in back testing using historical data. This highlights the urgent need for a more advanced methodology capable of addressing these challenges. There is a significant disconnect in the effective utilization of financial indicators to better understand the potential market trends of individual securities. The disclosure of successful trading strategies is often restricted within financial markets, resulting in a scarcity of widely documented and published strategies leveraging RL. Furthermore, current research frequently overlooks the identification of financial indicators correlated with various market trends and their potential advantages. This research endeavors to address these complexities by enhancing the ability of RL agents to effectively differentiate between positive and negative buy/sell actions using financial indicators. While we do not address all concerns, this paper provides deeper insights and commentary on the utilization of technical indicators and their benefits within reinforcement learning. This work establishes a foundational framework for further exploration and investigation of more complex scenarios.

q-fin.TR↗

arXiv2023

Graph Neural Operators for Classification of Spatial Transcriptomics Data

The inception of spatial transcriptomics has allowed improved comprehension of tissue architectures and the disentanglement of complex underlying biological, physiological, and pathological processes through their positional contexts. Recently, these contexts, and by extension the field, have seen much promise and elucidation with the application of graph learning approaches. In particular, neural operators have risen in regards to learning the mapping between infinite-dimensional function spaces. With basic to deep neural network architectures being data-driven, i.e. dependent on quality data for prediction, neural operators provide robustness by offering generalization among different resolutions despite low quality data. Graph neural operators are a variant that utilize graph networks to learn this mapping between function spaces. The aim of this research is to identify robust machine learning architectures that integrate spatial information to predict tissue types. Under this notion, we propose a study incorporating various graph neural network approaches to validate the efficacy of applying neural operators towards prediction of brain regions in mouse brain tissue samples as a proof of concept towards our purpose. We were able to achieve an F1 score of nearly 72% for the graph neural operator approach which outperformed all baseline and other graph network approaches.

cs.LG↗

arXiv2020

A Machine Learning Approach for Increased Throughput of Density Functional Theory Substitutional Alloy Studies

In this study, a machine learning-based technique is developed to reduce the computational cost required to explore large design spaces of substitutional alloys. The first advancement is based on a neural network approach to predict the initial position of ions for both minority and majority ions before ion relaxation. The second advancement is to allow the neural network to predict the total energy for every possibility minority ion position and select the most stable configuration in the absence of relaxing each trial position. In this study, a bismuth oxide materials system, (Bi$_{x}$La$_{y}$Yb$_{z}$)$_2$ MoO$_6$, is used as the model system to demonstrate the developed method and potential computational speedup. Comparing a brute force method that requires the calculation of every possible minority concentration location and subsequent relaxation there was a 1.3x speedup if the neural network (NN) was allowed to predict the initial position prior to relaxation. This speedup is a result in an average decrease of 4 wall hour (64 cpu-hrs) reduction in relaxation for individual calculations. Implementation of the second advancement allowed the NN to predict the total energy for all possible trials prior to relaxation, resulting in a speedup of approximately 37x. Validation was done by comparing both position and energy between the NN to DFT calculations. A maximum vector mean squared error (MSE) of 1.6x10$^{-2}$ and a maximum energy MSE of 2.3x10$^{-7}$ was predicted. This method demonstrates a significant computational speedup, which has the potential for significant computational savings for larger compositional design spaces.

physics.comp-ph↗

arXiv2018

Density Functional Theory Evaluation of Cation-doped Bismuth Molybdenum Oxide Photocatalysts for Nitrogen Fixation

This study investigates the photocatalytic nitrogen fixation on a cation-doped surface (Bi$_{x}$M$_{y}$)$_2$MoO$_6$ where (M = Fe, La, Yb) in both the orthorhombic and monoclinic configurations using a density functional theory (DFT) approach with experimentally validated model inputs. The proceeding discussion focuses on the Heyrovsky-type reactions for both the associative and dissociative reaction pathway related to nitrogen reduction. Key fundamental insight in the reduction mechanism is discussed that relates the material properties of the substitutional ions to the nitrogen and hydrogen affinities. Physical insight is gathered through interpretation of bound electronic states at the surface. Compositional phases of higher Fe and Yb concentrations resulted in decreased Mo-O binding and increased affinity between Mo and the N and H species on the surface. The modulation of the Mo-O binding is induced by strain as Yb and Fe are implemented, this, in turn, shifts energy levels and modulates the band gap energy by approximately 0.2 eV. This modification of Mo-O bond as substitution occurs is a result of the orbital hybridization of M-O (M = Fe, Yb) that causes a strong orbital interaction that shifts states. The optimal composition was predicted to be an orthorhombic configuration of (Bi$_{0.75}$Fe$_{0.25}$)$_2$MoO$_6$ with a predicted maximum thermodynamic energy barrier of 1.4 eV. This composition demonstrates effective nitrogen and hydrogen affinity that follows the associative or biological nitrogen fixation pathway.

cond-mat.mtrl-sci↗

arXiv2017

Ab-initio Study of the Electron Mobility in a Functionalized UiO-66 Metal Organic Framework

This study leverages density function theory (DFT) accompanied with Boltzmann transport equation approaches to investigate the electronic mobility as a function of inorganic substitution and functionalization in a thermally stable UiO-66 metal organic framework (MOF). The MOFs investigated are based on Zr-UiO-66 MOF with three functionalization groups of benzene dicarboxylate (BDC), BDC functionalized with an amino group (BDC + NH$_2$) and a nitro group (BDC + NO$_2$). The design space of this study is bound by UiO-66(M)-R, [M=Zr, Ti, Hf; R=BDC, BDC+NO$_2$, BDC+NH$_2$]. The elastic modulus was not found to vary significantly over the structural modification of the design space for either functionalization and inorganic substitution. However, the electron-phonon scattering potential was found to be controllable by up to 30\% through controlled inorganic substitution in the metal clusters of the MOF structure. The highest electron mobility was predicted for a UiO-66(Hf$_5$Zr$_1$) achieving a value of approximately 1.4x10$^{-3}$ cm$^2$/V-s. It was determined that functionalization provides a controlled method of modulating the charge density, while inorganic substitution provides a controlled method of modulating the electronic mobility. Within the proposed design space the electrical conductivity was able to be increased by approximately three times the base conductivity through a combination of inorganic substitution and functionalization.

cond-mat.mtrl-sci↗

arXiv2017

Ab initio Screening of a Sulfur Desorbed MoS$_2$ Photocatalyst for Nitrogen Fixation

The following study investigates the thermodynamic reaction barriers during nitrogen fixation for an inorganic sulfur desorbed photocatalyst Molybdenum disulfide surface. The design space is investigated using an density functional theory (DFT) method within a space bound by MMoS$_2$ M=Mo,Fe,Co. The discussion focuses on Heyrovsky type reactions along both the associative and dissociative pathway. A key insight into the roles of the inorganic and the balance between nitrogen and hydrogen affinity, providing evidence for an optimal material that minimizes the required over-potential. It is found that phases with a higher concentration of Mo face high reaction barrier involving nitrogen, where phases with higher concentrations of Fe and Co face high reaction barriers involving hydrogen species. In the absence of kinetic considerations, the best phase was predicted to be the 1T phase with a Mo$_{0.75}$Fe$_{0.25}$S$_{2}$ composition. This phase proved to have a balance of hydrogen and nitrogen affinity and follows the dissociative pathway, which can be evolved through non-thermal methods.

physics.chem-ph↗