arXiv subjects

Zahra Ahmadi

Publications and source records attributed to Zahra Ahmadi.

30 records · Page 2Linked to original sources

arXiv2024

Towards Precision Healthcare: Robust Fusion of Time Series and Image Data

With the increasing availability of diverse data types, particularly images and time series data from medical experiments, there is a growing demand for techniques designed to combine various modalities of data effectively. Our motivation comes from the important areas of predicting mortality and phenotyping where using different modalities of data could significantly improve our ability to predict. To tackle this challenge, we introduce a new method that uses two separate encoders, one for each type of data, allowing the model to understand complex patterns in both visual and time-based information. Apart from the technical challenges, our goal is to make the predictive model more robust in noisy conditions and perform better than current methods. We also deal with imbalanced datasets and use an uncertainty loss function, yielding improved results while simultaneously providing a principled means of modeling uncertainty. Additionally, we include attention mechanisms to fuse different modalities, allowing the model to focus on what's important for each task. We tested our approach using the comprehensive multimodal MIMIC dataset, combining MIMIC-IV and MIMIC-CXR datasets. Our experiments show that our method is effective in improving multimodal deep learning for clinical applications. The code will be made available online.

eess.IV↗

arXiv2024

A process mining-based error correction approach to improve data quality of an IoT-sourced event log

Internet of Things (IoT) systems are vulnerable to data collection errors and these errors can significantly degrade the quality of collected data, impact data analysis and lead to inaccurate or distorted results. This article emphasizes the importance of evaluating data quality and errors before proceeding with analysis and considering the effectiveness of error correction methods for a smart home use case.

cs.DB↗

arXiv2023

MTS2Graph: Interpretable Multivariate Time Series Classification with Temporal Evolving Graphs

Conventional time series classification approaches based on bags of patterns or shapelets face significant challenges in dealing with a vast amount of feature candidates from high-dimensional multivariate data. In contrast, deep neural networks can learn low-dimensional features efficiently, and in particular, Convolutional Neural Networks (CNN) have shown promising results in classifying Multivariate Time Series (MTS) data. A key factor in the success of deep neural networks is this astonishing expressive power. However, this power comes at the cost of complex, black-boxed models, conflicting with the goals of building reliable and human-understandable models. An essential criterion in understanding such predictive deep models involves quantifying the contribution of time-varying input variables to the classification. Hence, in this work, we introduce a new framework for interpreting multivariate time series data by extracting and clustering the input representative patterns that highly activate CNN neurons. This way, we identify each signal's role and dependencies, considering all possible combinations of signals in the MTS input. Then, we construct a graph that captures the temporal relationship between the extracted patterns for each layer. An effective graph merging strategy finds the connection of each node to the previous layer's nodes. Finally, a graph embedding algorithm generates new representations of the created interpretable time-series features. To evaluate the performance of our proposed framework, we run extensive experiments on eight datasets of the UCR/UEA archive, along with HAR and PAM datasets. The experiments indicate the benefit of our time-aware graph-based representation in MTS classification while enriching them with more interpretability.

cs.LG↗

arXiv2022

Steering line waves at a dual metasurface for optical applications

Line waves are defined as confined edge modes propagating at the interface of dual electromagnetic metasurfaces that preserve mirror reflection symmetries. Previous works have theoretically and practically explored these waves, showing that they occur at microwave regimes and terahertz ranges. It is also demonstrated that line waves can happen when there is a symmetric resistance discontinuity from negative to positive values and a uniform surface reactance. Line waves are of tunable mode confinement, direction-dependent polarizations, and singular field enhancement. This study presents a graphene patch design and demonstrates that the line waves associated with this structure can travel in the optical domain. Our design consists of a graphene metasurface on an epsilon near zero (ENZ) substrate. While our approach considers both dual reactive impedances and homogenous reactive ones, we only concentrate on line wave utilization in a dual-impedance structure.

physics.optics↗

arXiv2022

Physical Vapor Transport Growth of Antiferromagnetic CrCl$_3$ Flakes Down to Monolayer Thickness

The van der Waals magnets CrX$_3$ (X = I, Br, and Cl) exhibit highly tunable magnetic properties and are promising candidates for developing novel two-dimensional (2D) magnetic devices such as magnetic tunnel junctions and spin tunneling transistors. Previous studies of CrCl$_3$ have mainly focused on mechanically exfoliated samples. Controlled synthesis of high quality atomically thin flakes is critical for their technological implementation but has not been achieved to date. Here, we report the growth of large CrCl$_3$ flakes with well-defined facets down to monolayer thickness (~0.6 nm) via the physical vapor transport technique. Both isolated flakes with well-defined facets and long stripe samples with the trilayer portion exceeding 60 $μ$m have been obtained. High-resolution transmission electron microscopy studies show that the CrCl$_3$ flakes are single crystalline in the monoclinic structure, consistent with the Raman results. The room temperature stability of the CrCl$_3$ flakes decreases with decreasing thickness. The tunneling magnetoresistance of graphite/CrCl$_3$/graphite tunnel junctions confirms that few-layer CrCl$_3$ possesses in-plane magnetic anisotropy and Néel temperature of 17 K. Our study paves the path for developing CrCl$_3$-based scalable 2D spintronic applications.

cond-mat.mes-hall↗

arXiv2022

MANDO: Multi-Level Heterogeneous Graph Embeddings for Fine-Grained Detection of Smart Contract Vulnerabilities

Learning heterogeneous graphs consisting of different types of nodes and edges enhances the results of homogeneous graph techniques. An interesting example of such graphs is control-flow graphs representing possible software code execution flows. As such graphs represent more semantic information of code, developing techniques and tools for such graphs can be highly beneficial for detecting vulnerabilities in software for its reliability. However, existing heterogeneous graph techniques are still insufficient in handling complex graphs where the number of different types of nodes and edges is large and variable. This paper concentrates on the Ethereum smart contracts as a sample of software codes represented by heterogeneous contract graphs built upon both control-flow graphs and call graphs containing different types of nodes and links. We propose MANDO, a new heterogeneous graph representation to learn such heterogeneous contract graphs' structures. MANDO extracts customized metapaths, which compose relational connections between different types of nodes and their neighbors. Moreover, it develops a multi-metapath heterogeneous graph attention network to learn multi-level embeddings of different types of nodes and their metapaths in the heterogeneous contract graphs, which can capture the code semantics of smart contracts more accurately and facilitate both fine-grained line-level and coarse-grained contract-level vulnerability detection. Our extensive evaluation of large smart contract datasets shows that MANDO improves the vulnerability detection results of other techniques at the coarse-grained contract level. More importantly, it is the first learning-based approach capable of identifying vulnerabilities at the fine-grained line-level, and significantly improves the traditional code analysis-based vulnerability detection approaches by 11.35% to 70.81% in terms of F1-score.

cs.SE↗

arXiv2022

Quantum Diffusion in Sharp Transition to Non-Slow-Roll Phase

Transitions between different inflationary slow-roll scenarios are known to provide short non-slow-roll periods with non-trivial consequences. We consider the effect of quantum diffusion on the inflationary dynamics in a transition process. Using the stochastic δN formalism, we follow the detailed evolution of noises through a sharp transition modeled by the Starobinsky potential, although some of our results apply to any sharp transition. We find how the stochastic noise induced by the transition affects the coarse-grained fields. We then consider the special case that the potential is flat after the transition. It is found that the particular noise we obtain cannot drive the inflaton past the classically unreachable field values. By deriving the characteristic function, we also study the tail behavior for the distribution of curvature perturbations ζ, which we find to decay faster than e^(-3ζ).

gr-qc↗

arXiv2021

Intrinsic ferroelectricity in Y-doped HfO2 thin films

Ferroelectric HfO2-based materials hold great potential for widespread integration of ferroelectricity into modern electronics due to their robust ferroelectric properties at the nanoscale and compatibility with the existing Si technology. Earlier work indicated that the nanometer crystal grain size was crucial for stabilization of the ferroelectric phase of hafnia. This constraint caused high density of unavoidable structural defects of the HfO2-based ferroelectrics, obscuring the intrinsic ferroelectricity inherited from the crystal space group of bulk HfO2. Here, we demonstrate the intrinsic ferroelectricity in Y-doped HfO2 films of high crystallinity. Contrary to the common expectation, we show that in the 5% Y-doped HfO2 epitaxial thin films, high crystallinity enhances the spontaneous polarization up to a record-high 50 μC/cm2 value at room temperature. The high spontaneous polarization persists at reduced temperature, with polarization values consistent with our theoretical predictions, indicating the dominant contribution from the intrinsic ferroelectricity. The crystal structure of these films reveals the Pca21 orthorhombic phase with a small rhombohedral distortion, underlining the role of the anisotropic stress and strain. These results open a pathway to controlling the intrinsic ferroelectricity in the HfO2-based materials and optimizing their performance in applications.

cond-mat.mtrl-sci↗

arXiv2021

Focusing Knowledge-based Graph Argument Mining via Topic Modeling

Decision-making usually takes five steps: identifying the problem, collecting data, extracting evidence, identifying pro and con arguments, and making decisions. Focusing on extracting evidence, this paper presents a hybrid model that combines latent Dirichlet allocation and word embeddings to obtain external knowledge from structured and unstructured data. We study the task of sentence-level argument mining, as arguments mostly require some degree of world knowledge to be identified and understood. Given a topic and a sentence, the goal is to classify whether a sentence represents an argument in regard to the topic. We use a topic model to extract topic- and sentence-specific evidence from the structured knowledge base Wikidata, building a graph based on the cosine similarity between the entity word vectors of Wikidata and the vector of the given sentence. Also, we build a second graph based on topic-specific articles found via Google to tackle the general incompleteness of structured knowledge bases. Combining these graphs, we obtain a graph-based model which, as our evaluation shows, successfully capitalizes on both structured and unstructured data.

cs.IR↗

arXiv2020

Rule Extraction from Binary Neural Networks with Convolutional Rules for Model Validation

Most deep neural networks are considered to be black boxes, meaning their output is hard to interpret. In contrast, logical expressions are considered to be more comprehensible since they use symbols that are semantically close to natural language instead of distributed representations. However, for high-dimensional input data such as images, the individual symbols, i.e. pixels, are not easily interpretable. We introduce the concept of first-order convolutional rules, which are logical rules that can be extracted using a convolutional neural network (CNN), and whose complexity depends on the size of the convolutional filter and not on the dimensionality of the input. Our approach is based on rule extraction from binary neural networks with stochastic local search. We show how to extract rules that are not necessarily short, but characteristic of the input, and easy to visualize. Our experiments show that the proposed approach is able to model the functionality of the neural network while at the same time producing interpretable logical rules.

cs.LG↗

arXiv2018

Online Multi-Label Classification: A Label Compression Method

Many modern applications deal with multi-label data, such as functional categorizations of genes, image labeling and text categorization. Classification of such data with a large number of labels and latent dependencies among them is a challenging task, and it becomes even more challenging when the data is received online and in chunks. Many of the current multi-label classification methods require a lot of time and memory, which make them infeasible for practical real-world applications. In this paper, we propose a fast linear label space dimension reduction method that transforms the labels into a reduced encoded space and trains models on the obtained pseudo labels. Additionally, it provides an analytical method to update the decoding matrix which maps the labels into the original space and is used during the test phase. Experimental results show the effectiveness of this approach in terms of running times and the prediction performance over different measures.

cs.LG↗

arXiv2016

Room Temperature Ferroelectricity in Continuous Croconic Acid Thin Films

Ferroelectricity at room temperature has been demonstrated in nanometer-thin quasi 2D croconic acid thin films, by the polarization hysteresis loop measurements in macroscopic capacitor geometry, along with observation and manipulation of the nanoscale domain structure by piezoresponse force microscopy. The fabrication of continuous thin films of the hydrogen-bonded croconic acid was achieved by the suppression of the thermal decomposition using low evaporation temperatures in high vacuum, combined with growth conditions far from thermal equilibrium. For nominal coverages >=20 nm, quasi 2D and polycrystalline films, with an average grain size of 50-100 nm and 3.5 nm roughness, can be obtained. Spontaneous ferroelectric domain structures of the thin films have been observed and appear to correlate with the grain patterns. The application of this solvent-free growth protocol may be a key to the development of flexible organic ferroelectric thin films for electronic applications.

cond-mat.mtrl-sci↗