The growing need for artificial intelligence in multitude of domains has over the past few years triggered some concerns related to latency and reliability when huge amounts of sensor data are moved from sensors to the cloud for computation. These ...

Conceptual-level explanations of ECG signals has been of great interest in order that arrhythmias can be automatically classified and explained. We leverage and evaluate the power of large language models (LLMs), specifically ChatGPT, to generate the ...

Federated Learning (FL) is a distributed machine learning (ML) approach that allows multiple devices to train a model while maintaining training data privately on edge devices. Enhancing model generalizability necessitates an increase in information for ...

This study addresses the challenging 3D bin packing problem (3D-BPP) using state-of-the-art algorithms and GPU acceleration. The focus is on enhancing memory-bound packing operations and compute-intensive Deep Q-network using CUDA kernels, resulting in ...

This paper implements and analyzes multiple networks with the goal of understanding their suitability for edge device applications such as X-ray threat detection. In this study, we use the state-of-the-art YOLO object detection model to solve this task ...

Consumer-grade depth sensors provide low-resolution depth maps; however, a high-resolution RGB camera is usually mounted on the same device and acquires a high-resolution image of the same scene. While deep learning and guided filtering methods gave ...

Suicide is a serious global issue taking 800,000 lives every year. AI-assisted early detection of suicide risk and stress levels using a user’s activity on online forums and social media websites has shown great promise previously. In addition to this, ...

While Generative Artificial Intelligence (GenAI) has made significant strides in various sectors, there’s a discernible gap in literature regarding its specific applications and implications within the Financial Technology (FinTech) landscape. This ...

The availability of “AI cameras" with both image sensors and tiny, in situ computing platforms, has opened up the possibility of running the first level vision analytics on the network edge. Vision-based edge applications, such as Automatic Face ...

Vision-based hand gesture recognition in human-computer interface design has useful applications in virtual-reality, gaming control, communication through sign language, medical rehabilitation etc. In many scenarios, such applications are deployed on ...

Deep Neural Networks (DNNs) have drawn attention because of their outstanding performance on various tasks. However, deploying full-fledged DNNs in resource-constrained devices (edge, mobile, IoT) is difficult due to their large size. To overcome the ...

Large Language Models (LLM) have gained huge popularity recently due to their problem-solving capability in multiple domains. Technically LLMs can be considered a critical mixture of huge amounts of training data, smart and exhaustive prompt engineering, ...

As the adoption of point prediction models, such as deep neural networks and Gradient Boosting methods, continues to grow in critical decision-making systems, ensuring the reliability of their inferences has become a paramount concern. These ...

The importance of the robustness of deep neural networks (DNNs) cannot be overemphasised due to their widespread use in safety-critical applications. However, the fragility of DNNs to imperceptible input perturbations is well-known in the literature. ...

Training machine and deep learning models directly on extremely resource-constrained devices is the next challenge in the field of tiny machine learning. The related literature in this field is very limited, since most of the solutions focus only on on-...

Functional MRI (fMRI) is widely used to examine brain functionality by detecting alteration in oxygenated blood flow that arises with brain activity. This work aims to investigate the neurological variation of human brain responses during viewing of ...

Recent advances in pre-trained neural language models have substantially enhanced the performance of numerous natural language processing (NLP) tasks. However, some existing models require pretraining on a large dataset. Moreover, on using a deep ...