Search

African American English (AAE) presents unique challenges in natural language processing (NLP). This research systematically compares the performance of available NLP models--rule-based, transformer-based, and large language models (LLMs)--capable of identifying key grammatical features of AAE, namely Habitual Be and Multiple Negation. These features were selected for their distinct grammatical complexity and frequency of occurrence. The evaluation involved sentence-level binary classification tasks, using both zero-shot and few-shot strategies. The analysis reveals that while LLMs show promise compared to the baseline, they are influenced by biases such as recency and unrelated features in the text such as formality. This study highlights the necessity for improved model training and architectural adjustments to better accommodate AAE's unique linguistic characteristics. Data and code are available., Comment: 13 pages, Accepted to "Findings of the Association for Computational Linguistics: NAACL 2025"

Nontrivial topological invariant of bulk electronic wavefunctions in two-dimensional quantum crystals leaves its footprints on the edge, dislocation, and corner modes. Here we investigate non-unitary time dynamics of these topological modes in square lattice-based open quantum systems in which the time-dependent Hamiltonian smoothly interpolates between topologically distinct insulators across band gap closing quantum critical points. The temporal dynamics of these modes is described by a Lindblad equation in which the instantaneous Hamiltonian plays the role of the Lindblad operator, thereby allowing the environment to couple with the system through the energy channels (weak measurement protocol). We show that in the presence of such a real time ramp, the survival probability of these modes decreases (increases) in short (long) time scale where the dephasing (quantum Zeno) effect dominates with the increasing amplitude of the system-to-environment coupling, for both slow and fast ramps from a topological to a normal insulating state. For a reverse course of the time evolution, the revival or condensation probability of nucleating such topological modes, otherwise absent in the initial system, increases for stronger system-to-environment coupling. This phenomenon can be attributed to the strong decoherence of the initial mixed state among all the energy eigenstates of the final Hamiltonian which also includes the topological modes, causing their enhanced condensation probability. Our findings can be germane to real open topological materials with time-tunable band gap, and should be applicable to open topological crystals of arbitrary dimension and belonging to any symmetry class., Comment: 18 Pages, 17 Figures

Automated visualization recommendations (vis-rec) help users to derive crucial insights from new datasets. Typically, such automated vis-rec models first calculate a large number of statistics from the datasets and then use machine-learning models to score or classify multiple visualizations choices to recommend the most effective ones, as per the statistics. However, state-of-the art models rely on very large number of expensive statistics and therefore using such models on large datasets become infeasible due to prohibitively large computational time, limiting the effectiveness of such techniques to most real world complex and large datasets. In this paper, we propose a novel reinforcement-learning (RL) based framework that takes a given vis-rec model and a time-budget from the user and identifies the best set of input statistics that would be most effective while generating the visual insights within a given time budget, using the given model. Using two state-of-the-art vis-rec models applied on three large real-world datasets, we show the effectiveness of our technique in significantly reducing time-to visualize with very small amount of introduced error. Our approach is about 10X times faster compared to the baseline approaches that introduce similar amounts of error., Comment: Accepted at PAKDD 2024 (Oral)

This paper introduces Dirichlet process mixtures of block $g$ priors for model selection and prediction in linear models. These priors are extensions of traditional mixtures of $g$ priors that allow for differential shrinkage for various (data-selected) blocks of parameters while fully accounting for the predictors' correlation structure, providing a bridge between the literatures on model selection and continuous shrinkage priors. We show that Dirichlet process mixtures of block $g$ priors are consistent in various senses and, in particular, that they avoid the conditional Lindley ``paradox'' highlighted by Som et al.(2016). Further, we develop a Markov chain Monte Carlo algorithm for posterior inference that requires only minimal ad-hoc tuning. Finally, we investigate the empirical performance of the prior in various real and simulated datasets. In the presence of a small number of very large effects, Dirichlet process mixtures of block $g$ priors lead to higher power for detecting smaller but significant effects without only a minimal increase in the number of false discoveries.

This paper will illustrate the usage of Machine Learning algorithms on US College Scorecard datasets. For this paper, we will use our knowledge, research, and development of a predictive model to compare the results of all the models and predict the public and private net prices. This paper focuses on analyzing US College Scorecard data from data published on government websites. Our goal is to use four machine learning regression models to develop a predictive model to forecast the equitable net cost for every college, encompassing both public institutions and private, whether for-profit or nonprofit., Comment: 4 pages, 1 figure, 5 Tables

In this article, a posteriori error analysis of the elliptic obstacle problem is addressed using hybrid high-order methods. The method involve cell unknowns represented by degree-$r$ polynomials and face unknowns represented by degree-$s$ polynomials, where $r=0$ and $s$ is either $0$ or $1$. The discrete obstacle constraints are specifically applied to the cell unknowns. The analysis hinges on the construction of a suitable Lagrange multiplier, a residual functional and a linear averaging map. The reliability and the efficiency of the proposed a posteriori error estimator is discussed, and the study is concluded by numerical experiments supporting the theoretical results.

Data generated by edge devices has the potential to train intelligent autonomous systems across various domains. Despite the emergence of diverse machine learning approaches addressing privacy concerns and utilizing distributed data, security issues persist due to the sensitive storage of data shards in disparate locations. This paper introduces a potentially groundbreaking paradigm for machine learning model training, specifically designed for scenarios with only a single magnetic image and its corresponding label image available. We harness the capabilities of Deep Learning to generate concise yet informative samples, aiming to overcome data scarcity. Through the utilization of deep learning's internal representations, our objective is to efficiently address data scarcity issues and produce meaningful results. This methodology presents a promising avenue for training machine learning models with minimal data., Comment: 5 Pages, 6 Figures

We present efficient MATLAB implementations of the lowest-order primal hybrid finite element method (FEM) for linear second-order elliptic and parabolic problems with mixed boundary conditions in two spatial dimensions. We employ the Crank-Nicolson finite difference scheme for the complete discrete setup of the parabolic problem. All the codes presented are fully vectorized using matrix-wise array operations. Numerical experiments are conducted to show the performance of the software.

In this paper, we develop a new residual-based pointwise a posteriori error estimator of the quadratic finite element method for the Signorini problem. The supremum norm a posteriori error estimates enable us to locate the singularities locally to control the pointwise errors. In the analysis the discrete counterpart of contact force density is constructed suitably to exhibit the desired sign property. We employ a priori estimates for the standard Green's matrix for the divergence type operator and introduce the upper and lower barriers functions by appropriately modifying the discrete solution. Finally, we present numerical experiments that illustrate the excellent performance of the proposed error estimator.

An a posteriori error bound for the pointwise error of the quadratic discontinuous Galerkin method for the unilateral contact problem on polygonal domain is presented. The pointwise a posteriori error analysis is based on the direct use of a priori estimates of the Green's matrix for the divergence type operators and the suitable construction of the discrete contact force density $\b{\sigma}_h$ and barrier functions for the continuous solution. Several numerical experiments (in two dimension) are presented to illustrate the reliability and efficiency properties of the proposed aposteriori error estimator.

In this article, we addressed the numerical solution of a non-linear evolutionary variational inequality, which is encountered in the investigation of quasi-static contact problems. Our study encompasses both the semi-discrete and fully-discrete schemes, where we employ the backward Euler method for time discretization and utilize the lowest order Crouzeix-Raviart nonconforming finite element method for spatial discretization. By assuming appropriate regularity conditions on the solution, we establish \emph{a priori} error analysis for these schemes, achieving the optimal convergence order for linear elements. To illustrate the numerical convergence rates, we provide numerical results on a two-dimensional test problem.

In this article, we employ discontinuous Galerkin (DG) methods for the finite element approximation of the frictionless unilateral contact problem using quadratic finite elements over simplicial triangulation. We first establish an optimal \textit{a priori} error estimates under the appropriate regularity assumption on the exact solution $\b{u}$. Further, we analyze \textit{a posteriori} error estimates in the DG norm wherein, the reliability and efficiency of the proposed \textit{a posteriori} error estimator is addressed. The suitable construction of discrete Lagrange multiplier $\b{\lambda_h}$ and some intermediate operators play a key role in developing \textit{a posteriori} error analysis. Numerical results presented on uniform and adaptive meshes illustrate and confirm the theoretical findings.

Answering questions using pre-trained language models (LMs) and knowledge graphs (KGs) presents challenges in identifying relevant knowledge and performing joint reasoning.We compared LMs (fine-tuned for the task) with the previously published QAGNN method for the Question-answering (QA) objective and further measured the impact of additional factual context on the QAGNN performance. The QAGNN method employs LMs to encode QA context and estimate KG node importance, and effectively update the question choice entity representations using Graph Neural Networks (GNNs). We further experimented with enhancing the QA context encoding by incorporating relevant knowledge facts for the question stem. The models are trained on the OpenbookQA dataset, which contains ~6000 4-way multiple choice questions and is widely used as a benchmark for QA tasks. Through our experimentation, we found that incorporating knowledge facts context led to a significant improvement in performance. In contrast, the addition of knowledge graphs to language models resulted in only a modest increase. This suggests that the integration of contextual knowledge facts may be more impactful for enhancing question answering performance compared to solely adding knowledge graphs., Comment: 5 pages, 4 figures, 4 tables

Document Summarization is the procedure of generating a meaningful and concise summary of a given document with the inclusion of relevant and topic-important points. There are two approaches: one is picking up the most relevant statements from the document itself and adding it to the Summary known as Extractive and the other is generating sentences for the Summary known as Abstractive Summarization. Training a machine learning model to perform tasks that are time-consuming or very difficult for humans to evaluate is a major challenge. Book Abstract generation is one of such complex tasks. Traditional machine learning models are getting modified with pre-trained transformers. Transformer based language models trained in a self-supervised fashion are gaining a lot of attention; when fine-tuned for Natural Language Processing(NLP) downstream task like text summarization. This work is an attempt to use Transformer based techniques for Abstract generation., Comment: Published at - https://ijisae.org/index.php/IJISAE/article/view/2421

Machine learning has helped advance the field of anomaly detection by incorporating classifiers and autoencoders to decipher between normal and anomalous behavior. Additionally, federated learning has provided a way for a global model to be trained with multiple clients' data without requiring the client to directly share their data. This paper proposes a novel anomaly detector via federated learning to detect malicious network activity on a client's server. In our experiments, we use an autoencoder with a classifier in a federated learning framework to determine if the network activity is benign or malicious. By using our novel min-max scalar and sampling technique, called FedSam, we determined federated learning allows the global model to learn from each client's data and, in turn, provide a means for each client to improve their intrusion detection system's defense against cyber-attacks.

We perform a posteriori error analysis in the supremum norm for the quadratic discontinuous Galerkin method for the elliptic obstacle problem. We define two discrete sets (motivated by Gaddam, Gudi and Kamana [1]), one set having integral constraints and other one with the nodal constraints at the quadrature points, and discuss the pointwise reliability and efficiency of the proposed a posteriori error estimator. In the analysis, we employ a linear averaging function to transfer DG finite element space to standard conforming finite element space and exploit the sharp bounds on the Green's function of the Poisson's problem. Moreover, the upper and the lower barrier functions corresponding to continuous solution u are constructed by modifying the conforming part of the discrete solution uh appropriately. Finally, numerical experiments are presented to complement the theoretical results.

In this paper, we present and analyze a posteriori error estimates in the energy norm of a quadratic finite element method for the frictionless unilateral contact problem. The reliability and the efficiency of a posteriori error estimator is discussed. The suitable decomposition of the discrete space $\b{V^h}$ and a discrete space $\b{Q^h}$, where the discrete counterpart of the contact force density is defined, play crucial role in deriving a posteriori error estimates. Numerical results are presented exhibiting the reliability and the efficiency of the proposed error estimator.

The main aim of this article is to analyze mixed finite element method for the second order Dirichlet boundary control problem. Therein, we develop both a priori and a posteriori error analysis using the energy space based approach. We obtain optimal order a priori error estimates in the energy norm and $L^2$-norm with the help of auxiliary problems. The reliability and the efficiency of proposed a posteriori error estimator is discussed using the Helmholtz decomposition. Numerical experiments are presented to confirm the theoretical findings.

Due to the recent challenges in post-quantum cryptography, several new approaches for code-based cryptography have been proposed. For example, a variant of the McEliece cryptosystem based on interleaved codes was proposed. In order to deem such new settings secure, we first need to understand and analyze the complexity of the underlying problem, in this case the problem of decoding a random interleaved code. A simple approach to decode such codes, would be to randomly choose a vector in the row span of the received matrix and run a classical information set decoding algorithm on this erroneous codeword. In this paper, we propose a new generic decoder for interleaved codes, which is an adaption of the classical idea of information set decoding by Prange and perfectly fits the interleaved setting. We then analyze the cost of the new algorithm and a comparison to the simple approach described above shows the superiority of Interleaved Prange.

The goal of Approximate Query Processing (AQP) is to provide very fast but "accurate enough" results for costly aggregate queries thereby improving user experience in interactive exploration of large datasets. Recently proposed Machine-Learning based AQP techniques can provide very low latency as query execution only involves model inference as compared to traditional query processing on database clusters. However, with increase in the number of filtering predicates(WHERE clauses), the approximation error significantly increases for these methods. Analysts often use queries with a large number of predicates for insights discovery. Thus, maintaining low approximation error is important to prevent analysts from drawing misleading conclusions. In this paper, we propose ELECTRA, a predicate-aware AQP system that can answer analytics-style queries with a large number of predicates with much smaller approximation errors. ELECTRA uses a conditional generative model that learns the conditional distribution of the data and at runtime generates a small (~1000 rows) but representative sample, on which the query is executed to compute the approximate result. Our evaluations with four different baselines on three real-world datasets show that ELECTRA provides lower AQP error for large number of predicates compared to baselines., Comment: To appear in Proceedings of AAAI 2022

This article is concerned with the nonconforming finite element method for distributed elliptic optimal control problems with pointwise constraints on the control and gradient of the state variable. We reduce the minimization problem into a pure state constraint minimization problem. In this case, the solution of the minimization problem can be characterized as fourth-order elliptic variational inequalities of the first kind. To discretize the control problem we have used the bubble enriched Morley finite element method. To ensure the existence of the solution to discrete problems three bubble functions corresponding to the mean of the edge are added to the discrete space. We derive the error in the state variable in $H^2$-type energy norm. Numerical results are presented to illustrate our analytical findings., Comment: 16 pages, 1 figures