ARArunava Roy

Arunava Roy

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Arunava Roy has a total of 19 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    What should GAN in AI stand for?

    by Dipankar Dasgupta, Arunava Roy

    GANs were introduced as an AI framework where two learning models (a generator and a discriminator) compete in a zero-sum, two-player game. Initially, GANs were applied in adversarial machine learning such as deepfakes for images, audio, and videos used in digital media so the name Generative Adversarial Networks was coined. While early research on GANs focused on adversarial applications, a diverse range of non-adversarial GAN applications soon emerged. It is to be noted that the GAN formulation is inherently neutral, specific application makes it adversarial. Accordingly, we argue that an appropriate interpretation of GAN should be “Generative Associated Networks” since it tries to associate a generated instance to the real one by minimizing their divergence. This misnomer correction is needed as GANs (based on their defining characteristics) are generating synthetic data, hyperrealistic images, videos, and other transformative digital content. Their influence extends across domains such as healthcare, art, and data augmentation, making this technique useful in many AI applications. Given this evolution, we feel that GAN should stand for “Generative Associated Networks”, reflecting their broader role in generating and associating meaningful representations beyond adversarial learning.

  • Open Access

    DRD-GAN: A Novel Distributed Conditional Wasserstein Deep Convolutional Relativistic Discriminator GAN with Improved Convergence

    by Arunava Roy, Dipankar Dasgupta

    Generative Adversarial Network (GAN) exhibited significant capabilities in many applications including image enhancement and manipulation, language translation, generating images/videos from text, creating art and music, and so on. However, training GANs using large datasets remains highly computationally intensive for most of the standalone systems. Additionally, standalone GANs often exhibit poor synchronization between their generator and discriminator with unstable training, poor convergence along with a large number of mode collapses and vanishing/exploding gradients. Standalone GANs also failed to learn in a decentralized environment, where the data is distributed among several client machines. Some researchers have lately used the most prevalent decentralized setting available today, called Federated Learning (FL) to develop distributed-GAN strategies as the possible solutions, although their implementations mostly failed to address the above issues mainly because of: the training instability within the distributed components, which eventually leads to the poor synchronization among the generators and discriminators scattered over several machines. In this work, we developed a computationally inexpensive Wasserstein conditional Distributed Relativistic Discriminator-GAN or DRD-GAN to alleviate the above issues. DRD-GAN stabilizes its training (with non-convex losses) by keeping a global generator in the central server and relativistic discriminators in the local clients (one discriminator per client), and uses Wasserstein-1 for computing local and global losses. It eventually avoids mode collapses, vanishing/exploding gradients (both in the presence of iid and non-iid samples) and helps DRD-GAN to produce high-quality fake images. Apart from that, the sheer unavailability of a capable conditional distributed-GAN model has become another motivation behind the current work. Essentially, we revisited the FL paradigms, locating one discriminator per client, and a generator in the central server that aggregates the updates coming from multiple discriminators. Relativistic discriminators in the clients are trained on both iid and non-iid private data. We presented a detailed mathematical formulation of DRD-GAN and empirically evaluated our claims using CIFAR-10, MNIST, EuroSAT, and CelebA datasets.


  • Open Access

    Pitfalls of Generic Large Language Models (GLLMs) from reliability and security perspectives

    by Dipankar Dasgupta, Arunava Roy

    Generic Large Language Models (GLLMs) have grown popularity in many professions with limited or no technical knowledge. Larger and larger GLLMs are continuously being released with enhanced capabilities, promoting the abilities of these Generative AI at the grassroots level in businesses. These tools excel in text, image, and video generation (assembling, summarizing, translating) when proper queries and prompts are given; moreover, various augmentation of up-to-date knowledge bases, making these more efficient in providing current events. Practitioners and marketers showcase the benefits of GLLMs by demonstrating various use cases. However, the reliability of GLLMs' responses is questionable in certain scenarios, particularly due to issues like hallucinations, factual inaccuracies, and inappropriate or unrelated responses. Also there remain many open questions on data collection, privacy and ethical issues that need to be addressed. This study emphasizes the reliability and security aspects of GLLMs while recognizing significant benefits in a wide variety of applications. We also provide some insides of social impacts and future directions of AI/ML applications.


  • Open Access

    A Critical Analysis of Distributed-GANs: Approaches, Challenges and Future Directions

    by Arunava Roy, Dipankar Dasgupta

    Distributed Generative Adversarial Networks (D-GANs) are a recent addition to generative model literature, but they come with significant limitations and caveats. This paper categorizes and analyzes the existing D-GAN models and introduces a novel four-part categorization framework based on their mathematical properties. We provided generalized mathematical formulations for each D-GAN category, addressing convergence issues and potential improvements, which are essential for developing more robust and reliable D-GANs. Additionally, we conducted detailed mathematical analyses of existing D-GANs to identify their shortcomings in terms of expanse, convergence and output quality. This rigorous analysis is pivotal for advancing the field, as it highlights critical areas needing improvement and sets the stage for future innovations that will enhance the performance and reliability of distributed generative models. This work establishes a foundation for future research, and standardizes the evaluation of D-GAN models. Additionally, it guides the research community towards addressing unresolved issues and exploring new avenues, thereby enhancing the overall effectiveness and applicability of the distributed generative models in AI/ML.

  • A Distributed Conditional Wasserstein Deep Convolutional Relativistic Loss Generative Adversarial Network With Improved Convergence

    by Arunava Roy, Dipankar Dasgupta

    Generative Adversarial Networks (GAN) excel in diverse applications like image enhancement, manipulation, generating images and videos from text, etc. Yet, training GANs with large datasets remains computationally intensive for standalone systems. Synchronization issues between the generator and discriminator lead to unstable training, poor convergence, vanishing and exploding gradient challenges. In decentralized environments, standalone GANs struggle with distributed data on client machines. Researchers have turned to Federated Learning (FL) for distributed GAN implementations, but efforts often fall short due to training instability and poor synchronization within GAN components. In this study, we present DRL-GAN, a lightweight Wasserstein conditional Distributed Relativistic Loss-GAN designed to overcome existing limitations. DRL-GAN ensures training stability in the face of non-convex losses by employing a single global generator on the central server and a discriminator per client. Utilizing Wasserstein-1 for relativistic loss computation between real and fake samples, DRL-GAN effectively addresses issues like mode collapses, vanishing and exploding gradients, accommodating both iid and non-iid private data in clients and fostering strong convergence. The absence of a robust conditional distributed-GAN model serves as another motivation for this work. We provide a comprehensive mathematical formulation of DRL-GAN and validate our claims empirically on CIFAR-10, MNIST, EuroSAT, and LSUN-Bedroom datasets.


  • A Distributed Multi-User Access Control Middleware for Critical Applications

    by Alexander Williams, Arunava Roy, Dipankar Dasgupta

    We present a novel access control middleware for distributed multi-user review. The system uses a fuzzy inference system trained on real world access control rules to evaluate and select reviewers as an extension to a more traditional access control system. The method is intended for high security need specific requests, as a supplement to regular access control methods. In this way, it models a multi-person access system common in mechanical controls like missile launches, bank vault opening, and other high criticality domains. The proposed method improves security by increasing the number of compromised users needed to perform an attack, taking advantage of situational awareness of peer users in a system. We evaluate the proposed system with an example implementation based on a real-world organization, and show that the system can be used to effectively implement a secure resource access control system. Our work contributes to the growing body of research into fuzzy-logic access control, ML in access control, and multi-user authentication systems.

  • Open Access

    Clustering Cum Polar Coordinate Feature Transformation (C-PCFT) Approach to Identify Pores in Carbonate Rocks

    by Pradeep S. Naulia, Arunava Roy, Junzo Watada, Izzatdin Abdul Aziz

    Most of the world’s oil reserves and natural gas are stored within carbonate rock’s pores and fractures. Pores and fractures are quite popular for predicting the amount of petroleum under an adequate trap condition. Hence, their petrophysical properties, such as shape and size, are paramount for accurately predicting the reservoir’s state and condition. Current modelling techniques are mostly based on manual and expert judgement which are time-consuming and cost-intensive. In this study, we devised a robust and scalable image processing framework that uses the combination of pixel-based clustering approach with a polar coordinate feature transformation technique to intelligently identify the pores of carbonate rock samples. We reported that such a method can be effective in detecting pores of different shapes and sizes in an automated fashion. We rigorously tested the proposed method on the computed tomography-scanned micro-images of a carbonate rock sample, and the results demonstrate improved identification accuracy of the proposed method than the current deep learning counterparts. Another key advantage compared to deep learning methods, the proposed method does not require extensive training on data, which saves time and effort without being computationally too expensive.


  • A Novel Conditional Wasserstein Deep Convolutional Generative Adversarial Network

    by Arunava Roy, Dipankar Dasgupta

    Generative Adversarial Networks (GAN) and their several variants have not only been used for adversarial purposes but also used for extending the learning coverage of different AI/ML models. Most of these variants are unconditional and do not have enough control over their outputs. Conditional GANs (CGANs) have the ability to control their outputs by conditioning their generator and discriminator with an auxiliary variable (such as class label s, and text description s). However, CGANs have several drawbacks such as unstable training , non-convergence and multiple mode collapse s like other unconditional basic GANs (where the discriminator s are classifier s). DCGANs, WGANs, and MMDGANs enforce significant improvements to stabilize the GAN training although have no control over their outputs. We developed a novel conditional Wasserstein GAN model, called CWGAN ( a.k.a RD-GAN named after the initials of the authors' surnames ) that stabilizes GAN training by replacing relatively unstable JS divergence with Wasserstein-1 distance while maintaining better control over its outputs. We have shown that the CWGAN can produce optimal generator s and discriminator s irrespective of the original and input noise data distributions. We presented a detailed formulation of CWGAN and highlighted its salient features along with proper justifications. We showed the CWGAN has a wide variety of adversarial applications including preparing fake images through a CWGAN-based deep generative hashing function and generating highly accurate user mouse trajectories for fooling any underlying mouse dynamics authentications (MDAs). We conducted detailed experiments using well-known benchmark datasets in support of our claims.


  • Attacking Mouse Dynamics Authentication Using Novel Wasserstein Conditional DCGAN

    by Arunava Roy, Koksheik Wong, Raphaël C.-W. Phan

    Behavioral biometrics is an emerging trend due to their cost-effectiveness and non-intrusive implementations that support remote access for user identification. This is the case especially in recent times of social distancing and working from home arrangements, where online attendance is the preferred option in contrast to physical presence. In this work, we explore the limitations of mouse dynamics authentication by impersonating legitimate user mouse action sequences. Specifically, towards that aim, we develop a novel generative WC-DCGAN model to generate highly accurate fake user action sequences. We apply our WC-DCGAN to this problem and show that it causes the target classifier can be tricked into identifying a fraudster as a legitimate user. WC-DCGAN has several benefits, including: achieving dominated convergence, hence implying the existence of solutions and optimal discriminator regardless of data and generator distributions; and acting as an unsupervised model for a fixed class label and generator. Experiments are conducted to verify these points. Subsequently, we analyzed the cause of misclassifications, and propose a novel mouse dynamics strategy that offers much tighter authentication with significant reductions in misclassification events.