arXiv subjects

Rishabh Gupta

Publications and source records attributed to Rishabh Gupta.

At least 37 records · Page 2Linked to original sources

arXiv2024

Selective Shot Learning for Code Explanation

Code explanation plays a crucial role in the software engineering domain, aiding developers in grasping code functionality efficiently. Recent work shows that the performance of LLMs for code explanation improves in a few-shot setting, especially when the few-shot examples are selected intelligently. State-of-the-art approaches for such Selective Shot Learning (SSL) include token-based and embedding-based methods. However, these SSL approaches have been evaluated on proprietary LLMs, without much exploration on open-source Code-LLMs. Additionally, these methods lack consideration for programming language syntax. To bridge these gaps, we present a comparative study and propose a novel SSL method (SSL_ner) that utilizes entity information for few-shot example selection. We present several insights and show the effectiveness of SSL_ner approach over state-of-the-art methods across two datasets. To the best of our knowledge, this is the first systematic benchmarking of open-source Code-LLMs while assessing the performances of the various few-shot examples selection approaches for the code explanation task.

cs.SE↗

arXiv2024

Time resolved quantum tomography in molecular spectroscopy by the Maximal Entropy Approach

Attosecond science offers unprecedented precision in probing the initial moments of chemical reactions, revealing the dynamics of molecular electrons that shape reaction pathways. A fundamental question emerges: what role, if any, do quantum coherences between molecular electron states play in photochemical reactions? Answering this question necessitates quantum tomography: the determination of the electronic density matrix from experimental data, where the off-diagonal elements represent these coherences. The Maximal Entropy (MaxEnt) based Quantum State Tomography (QST) approach offers unique advantages in studying molecular dynamics, particularly with partial tomographic data. Here, we explore the application of MaxEnt-based QST on photoexcited ammonia, necessitating the operator form of observables specific to the performed measurements. We present two methodologies for constructing these operators: one leveraging Molecular Angular Distribution Moments (MADMs) which accurately capture the orientation-dependent vibronic dynamics of molecules; and another utilizing Angular Momentum Coherence Operators to construct measurement operators for the full rovibronic density matrix in the symmetric top basis. A key revelation of our study is the direct link between Lagrange multipliers in the MaxEnt formalism and the unique set of MADMs. Furthermore, we achieve a groundbreaking milestone by constructing, for the first time, the entanglement entropy of the electronic subsystem: a metric that was previously inaccessible. The entropy vividly reveals and quantifies the effects of coupling between the excited electron and nuclear degrees of freedom. Consequently, our findings open new avenues for research in ultrafast molecular spectroscopy within the broader domain of quantum information science.

physics.chem-ph↗

arXiv2024

Deep Learning Based Named Entity Recognition Models for Recipes

Food touches our lives through various endeavors, including flavor, nourishment, health, and sustainability. Recipes are cultural capsules transmitted across generations via unstructured text. Automated protocols for recognizing named entities, the building blocks of recipe text, are of immense value for various applications ranging from information extraction to novel recipe generation. Named entity recognition is a technique for extracting information from unstructured or semi-structured data with known labels. Starting with manually-annotated data of 6,611 ingredient phrases, we created an augmented dataset of 26,445 phrases cumulatively. Simultaneously, we systematically cleaned and analyzed ingredient phrases from RecipeDB, the gold-standard recipe data repository, and annotated them using the Stanford NER. Based on the analysis, we sampled a subset of 88,526 phrases using a clustering-based approach while preserving the diversity to create the machine-annotated dataset. A thorough investigation of NER approaches on these three datasets involving statistical, fine-tuning of deep learning-based language models and few-shot prompting on large language models (LLMs) provides deep insights. We conclude that few-shot prompting on LLMs has abysmal performance, whereas the fine-tuned spaCy-transformer emerges as the best model with macro-F1 scores of 95.9%, 96.04%, and 95.71% for the manually-annotated, augmented, and machine-annotated datasets, respectively.

cs.CL↗

arXiv2024

Digital Twin Generators for Disease Modeling

A patient's digital twin is a computational model that describes the evolution of their health over time. Digital twins have the potential to revolutionize medicine by enabling individual-level computer simulations of human health, which can be used to conduct more efficient clinical trials or to recommend personalized treatment options. Due to the overwhelming complexity of human biology, machine learning approaches that leverage large datasets of historical patients' longitudinal health records to generate patients' digital twins are more tractable than potential mechanistic models. In this manuscript, we describe a neural network architecture that can learn conditional generative models of clinical trajectories, which we call Digital Twin Generators (DTGs), that can create digital twins of individual patients. We show that the same neural network architecture can be trained to generate accurate digital twins for patients across 13 different indications simply by changing the training set and tuning hyperparameters. By introducing a general purpose architecture, we aim to unlock the ability to scale machine learning approaches to larger datasets and across more indications so that a digital twin could be created for any patient in the world.

cs.LG↗

arXiv2024

Entropy corrected geometric Brownian motion

The geometric Brownian motion (GBM) is widely employed for modeling stochastic processes, yet its solutions are characterized by the log-normal distribution. This comprises predictive capabilities of GBM mainly in terms of forecasting applications. Here, entropy corrections to GBM are proposed to go beyond log-normality restrictions and better account for intricacies of real systems. It is shown that GBM solutions can be effectively refined by arguing that entropy is reduced when deterministic content of considered data increases. Notable improvements over conventional GBM are observed for several cases of non-log-normal distributions, ranging from a dice roll experiment to real world data.

physics.data-an↗

arXiv2024

Sparse Graph Representations for Procedural Instructional Documents

Computation of document similarity is a critical task in various NLP domains that has applications in deduplication, matching, and recommendation. Traditional approaches for document similarity computation include learning representations of documents and employing a similarity or a distance function over the embeddings. However, pairwise similarities and differences are not efficiently captured by individual representations. Graph representations such as Joint Concept Interaction Graph (JCIG) represent a pair of documents as a joint undirected weighted graph. JCIGs facilitate an interpretable representation of document pairs as a graph. However, JCIGs are undirected, and don't consider the sequential flow of sentences in documents. We propose two approaches to model document similarity by representing document pairs as a directed and sparse JCIG that incorporates sequential information. We propose two algorithms inspired by Supergenome Sorting and Hamiltonian Path that replace the undirected edges with directed edges. Our approach also sparsifies the graph to $O(n)$ edges from JCIG's worst case of $O(n^2)$. We show that our sparse directed graph model architecture consisting of a Siamese encoder and GCN achieves comparable results to the baseline on datasets not containing sequential information and beats the baseline by ten points on an instructional documents dataset containing sequential information.

cs.CL↗

arXiv2023

Data-driven decision-focused surrogate modeling

We introduce the concept of decision-focused surrogate modeling for solving computationally challenging nonlinear optimization problems in real-time settings. The proposed data-driven framework seeks to learn a simpler, e.g. convex, surrogate optimization model that is trained to minimize the decision prediction error, which is defined as the difference between the optimal solutions of the original and the surrogate optimization models. The learning problem, formulated as a bilevel program, can be viewed as a data-driven inverse optimization problem to which we apply a decomposition-based solution algorithm from previous work. We validate our framework through numerical experiments involving the optimization of common nonlinear chemical processes such as chemical reactors, heat exchanger networks, and material blending systems. We also present a detailed comparison of decision-focused surrogate modeling with standard data-driven surrogate modeling methods and demonstrate that our approach is significantly more data-efficient while producing simple surrogate models with high decision prediction accuracy.

math.OC↗

arXiv2023

Exploring Large Language Models for Code Explanation

Automating code documentation through explanatory text can prove highly beneficial in code understanding. Large Language Models (LLMs) have made remarkable strides in Natural Language Processing, especially within software engineering tasks such as code generation and code summarization. This study specifically delves into the task of generating natural-language summaries for code snippets, using various LLMs. The findings indicate that Code LLMs outperform their generic counterparts, and zero-shot methods yield superior results when dealing with datasets with dissimilar distributions between training and testing sets.

cs.SE↗

arXiv2023

An AI-Driven VM Threat Prediction Model for Multi-Risks Analysis-Based Cloud Cybersecurity

Cloud virtualization technology, ingrained with physical resource sharing, prompts cybersecurity threats on users' virtual machines (VM)s due to the presence of inevitable vulnerabilities on the offsite servers. Contrary to the existing works which concentrated on reducing resource sharing and encryption and decryption of data before transfer for improving cybersecurity which raises computational cost overhead, the proposed model operates diversely for efficiently serving the same purpose. This paper proposes a novel Multiple Risks Analysis based VM Threat Prediction Model (MR-TPM) to secure computational data and minimize adversary breaches by proactively estimating the VMs threats. It considers multiple cybersecurity risk factors associated with the configuration and management of VMs, along with analysis of users' behaviour. All these threat factors are quantified for the generation of respective risk score values and fed as input into a machine learning based classifier to estimate the probability of threat for each VM. The performance of MR-TPM is evaluated using benchmark Google Cluster and OpenNebula VM threat traces. The experimental results demonstrate that the proposed model efficiently computes the cybersecurity risks and learns the VM threat patterns from historical and live data samples. The deployment of MR-TPM with existing VM allocation policies reduces cybersecurity threats up to 88.9%.

cs.CR↗

arXiv2023

Physics inspired quantum simulation of resonating valence bond states -- a prototypical template for a spin-liquid ground state

Spin-liquids -- an emergent, exotic collective phase of matter -- have garnered enormous attention in recent years. While experimentally, many prospective candidates have been proposed and realized, theoretically modeling real materials that display such behavior may pose serious challenges due to the inherently high correlation content of emergent phases. Over the last few decades, the second-quantum revolution has been the harbinger of a novel computational paradigm capable of initiating a foundational evolution in computational physics. In this report, we strive to use the power of the latter to study a prototypical model -- a spin-$\frac{1}{2}$-unit cell of a Kagome anti-ferromagnet. Extended lattices of such unit cells are known to possess a magnetically disordered spin-liquid ground state. We employ robust classical numerical techniques like Density-Matrix Renormalization Group (DMRG) to identify the nature of the ground state through a matrix-product state (MPS) formulation. We subsequently use the gained insight to construct an auxillary hamiltonian with reduced measurables and also design an ansatz that is modular and gate efficient. With robust error-mitigation strategies, we are able to demonstrate that the said ansatz is capable of accurately representing the target ground state even on a real IBMQ backend within $1\%$ accuracy in energy. Since the protocol is linearly scaling $O(n)$ in the number of unit cells, gate requirements, and the number of measurements, it is straightforwardly extendable to larger Kagome lattices which can pave the way for efficient construction of spin-liquid ground states on a quantum device.

cond-mat.str-el↗

arXiv2023

Cryptography approach for Secure Outsourced Data Storage in Cloud Environment

A large amount of data and applications are migrated by researchers, stakeholders, academia, and business organizations to the cloud environment due to its large variety of services, which involve the least maintenance cost, maximum flexibility, and on-demand service for storage, computation, and data distribution intentions. Despite the various characteristics the cloud environment supports, it also faces many challenges. However, data users may not completely trust a cloud environment that is engaged by a third party. Every cloud user always has a prime concern, i.e., security. Numerous methods have been designed to solve the issue of data security during data storage, calculation, and sharing across stakeholders and users. Nevertheless, there is a lack of existing methods that tackle the issue of the security of data when it is stored in a cloud environment. This article presents a precise security method that has handled the security of data while it is being shared and stored in the cloud. These methods have been utilized to lessen security assaults and prevent unauthorized parties from accessing the actual data. The article is concluded with some limitations and recommendations for the future in terms of secure data retention and distribution.

cs.CR↗

arXiv2023

Counterspeeches up my sleeve! Intent Distribution Learning and Persistent Fusion for Intent-Conditioned Counterspeech Generation

Counterspeech has been demonstrated to be an efficacious approach for combating hate speech. While various conventional and controlled approaches have been studied in recent years to generate counterspeech, a counterspeech with a certain intent may not be sufficient in every scenario. Due to the complex and multifaceted nature of hate speech, utilizing multiple forms of counter-narratives with varying intents may be advantageous in different circumstances. In this paper, we explore intent-conditioned counterspeech generation. At first, we develop IntentCONAN, a diversified intent-specific counterspeech dataset with 6831 counterspeeches conditioned on five intents, i.e., informative, denouncing, question, positive, and humour. Subsequently, we propose QUARC, a two-stage framework for intent-conditioned counterspeech generation. QUARC leverages vector-quantized representations learned for each intent category along with PerFuMe, a novel fusion module to incorporate intent-specific information into the model. Our evaluation demonstrates that QUARC outperforms several baselines by an average of 10% across evaluation metrics. An extensive human evaluation supplements our hypothesis of better and more appropriate responses than comparative systems.

cs.CL↗

arXiv2023

Speaker Profiling in Multiparty Conversations

In conversational settings, individuals exhibit unique behaviors, rendering a one-size-fits-all approach insufficient for generating responses by dialogue agents. Although past studies have aimed to create personalized dialogue agents using speaker persona information, they have relied on the assumption that the speaker's persona is already provided. However, this assumption is not always valid, especially when it comes to chatbots utilized in industries like banking, hotel reservations, and airline bookings. This research paper aims to fill this gap by exploring the task of Speaker Profiling in Conversations (SPC). The primary objective of SPC is to produce a summary of persona characteristics for each individual speaker present in a dialogue. To accomplish this, we have divided the task into three subtasks: persona discovery, persona-type identification, and persona-value extraction. Given a dialogue, the first subtask aims to identify all utterances that contain persona information. Subsequently, the second task evaluates these utterances to identify the type of persona information they contain, while the third subtask identifies the specific persona values for each identified type. To address the task of SPC, we have curated a new dataset named SPICE, which comes with specific labels. We have evaluated various baselines on this dataset and benchmarked it with a new neural model, SPOT, which we introduce in this paper. Furthermore, we present a comprehensive analysis of SPOT, examining the limitations of individual modules both quantitatively and qualitatively.

cs.CL↗

arXiv2023

Dimension reduction and redundancy removal through successive Schmidt decompositions

Quantum computers are believed to have the ability to process huge data sizes which can be seen in machine learning applications. In these applications, the data in general is classical. Therefore, to process them on a quantum computer, there is a need for efficient methods which can be used to map classical data on quantum states in a concise manner. On the other hand, to verify the results of quantum computers and study quantum algorithms, we need to be able to approximate quantum operations into forms that are easier to simulate on classical computers with some errors. Motivated by these needs, in this paper we study the approximation of matrices and vectors by using their tensor products obtained through successive Schmidt decompositions. We show that data with distributions such as uniform, Poisson, exponential, or similar to these distributions can be approximated by using only a few terms which can be easily mapped onto quantum circuits. The examples include random data with different distributions, the Gram matrices of iris flower, handwritten digits, 20newsgroup, and labeled faces in the wild. And similarly, some quantum operations such as quantum Fourier transform and variational quantum circuits with a small depth also may be approximated with a few terms that are easier to simulate on classical computers. Furthermore, we show how the method can be used to simplify quantum Hamiltonians: In particular, we show the application to randomly generated transverse field Ising model Hamiltonians. The reduced Hamiltonians can be mapped into quantum circuits easily and therefore can be simulated more efficiently.

quant-ph↗

arXiv2023

Coherence and Diversity through Noise: Self-Supervised Paraphrase Generation via Structure-Aware Denoising

In this paper, we propose SCANING, an unsupervised framework for paraphrasing via controlled noise injection. We focus on the novel task of paraphrasing algebraic word problems having practical applications in online pedagogy as a means to reduce plagiarism as well as ensure understanding on the part of the student instead of rote memorization. This task is more complex than paraphrasing general-domain corpora due to the difficulty in preserving critical information for solution consistency of the paraphrased word problem, managing the increased length of the text and ensuring diversity in the generated paraphrase. Existing approaches fail to demonstrate adequate performance on at least one, if not all, of these facets, necessitating the need for a more comprehensive solution. To this end, we model the noising search space as a composition of contextual and syntactic aspects and sample noising functions consisting of either one or both aspects. This allows for learning a denoising function that operates over both aspects and produces semantically equivalent and syntactically diverse outputs through grounded noise injection. The denoising function serves as a foundation for learning a paraphrasing function which operates solely in the input-paraphrase space without carrying any direct dependency on noise. We demonstrate SCANING considerably improves performance in terms of both semantic preservation and producing diverse paraphrases through extensive automated and manual evaluation across 4 datasets.

cs.CL↗

arXiv2022

Hamiltonian learning from time dynamics using variational algorithms

The Hamiltonian of a quantum system governs the dynamics of the system via the Schrodinger equation. In this paper, the Hamiltonian is reconstructed in the Pauli basis using measurables on random states forming a time series dataset. The time propagation is implemented through Trotterization and optimized variationally with gradients computed on the quantum circuit. We validate our output by reproducing the dynamics of unseen observables on a randomly chosen state not used for the optimization. Unlike the existing techniques that try and exploit the structure/properties of the Hamiltonian, our scheme is general and provides freedom with regard to what observables or initial states can be used while still remaining efficient with regard to implementation. We extend our protocol to doing quantum state learning where we solve the reverse problem of doing state learning given time series data of observables generated against several Hamiltonian dynamics. We show results on Hamiltonians involving XX, ZZ couplings along with transverse field Ising Hamiltonians and propose an analytical method for the learning of Hamiltonians consisting of generators of the SU(3) group. This paper is likely to pave the way toward using Hamiltonian learning for time series prediction within the context of quantum machine learning algorithms.

quant-ph↗

arXiv2022

A Privacy-Preserving Model based on Differential Approach for Sensitive Data in Cloud Environment

A large amount of data and applications need to be shared with various parties and stakeholders in the cloud environment for storage, computation, and data utilization. Since a third party operates the cloud platform, owners cannot fully trust this environment. However, it has become a challenge to ensure privacy preservation when sharing data effectively among different parties. This paper proposes a novel model that partitions data into sensitive and non-sensitive parts, injects the noise into sensitive data, and performs classification tasks using k-anonymization, differential privacy, and machine learning approaches. It allows multiple owners to share their data in the cloud environment for various purposes. The model specifies communication protocol among involved multiple untrusted parties to process owners data. The proposed model preserves actual data by providing a robust mechanism. The experiments are performed over Heart Disease, Arrhythmia, Hepatitis, Indian-liver-patient, and Framingham datasets for Support Vector Machine, K-Nearest Neighbor, Random Forest, Naive Bayes, and Artificial Neural Network classifiers to compute the efficiency in terms of accuracy, precision, recall, and F1-score of the proposed model. The achieved results provide high accuracy, precision, recall, and F1-score up to 93.75%, 94.11%, 100%, and 87.99% and improvement up to 16%, 29%, 12%, and 11%, respectively, compared to previous works.

cs.CR↗

arXiv2022

Efficient Learning of Decision-Making Models: A Penalty Block Coordinate Descent Algorithm for Data-Driven Inverse Optimization

Decision-making problems are commonly formulated as optimization problems, which are then solved to make optimal decisions. In this work, we consider the inverse problem where we use prior decision data to uncover the underlying decision-making process in the form of a mathematical optimization model. This statistical learning problem is referred to as data-driven inverse optimization. We focus on problems where the underlying decision-making process is modeled as a convex optimization problem whose parameters are unknown. We formulate the inverse optimization problem as a bilevel program and propose an efficient block coordinate descent-based algorithm to solve large problem instances. Numerical experiments on synthetic datasets demonstrate the computational advantage of our method compared to standard commercial solvers. Moreover, the real-world utility of the proposed approach is highlighted through two realistic case studies in which we consider estimating risk preferences and learning local constraint parameters of agents in a multiplayer Nash bargaining game.

math.OC↗