Pradeep Subedi

Hanqiu Chen, Xuebin Yao, Pradeep Subedi et al.

Edge computing is a distributed computing paradigm that collects and processes data at or near the source of data generation. The on-device learning at edge relies on device-to-device wireless communication to facilitate real-time data sharing and collaborative decision-making among multiple devices. This significantly improves the adaptability of the edge computing system to the changing environments. However, as the scale of the edge computing system is getting larger, communication among devices is becoming the bottleneck because of the limited bandwidth of wireless communication leads to large data transfer latency. To reduce the amount of device-to-device data transmission and accelerate on-device learning, in this paper, we propose Residual-INR, a fog computing-based communication-efficient on-device learning framework by utilizing implicit neural representation (INR) to compress images/videos into neural network weights. Residual-INR enhances data transfer efficiency by collecting JPEG images from edge devices, compressing them into INR format at the fog node, and redistributing them for on-device learning. By using a smaller INR for full image encoding and a separate object INR for high-quality object region reconstruction through residual encoding, our technique can reduce the encoding redundancy while maintaining the object quality. Residual-INR is a promising solution for edge on-device learning because it reduces data transmission by up to 5.16 x across a network of 10 edge devices. It also facilitates CPU-free accelerated on-device learning, achieving up to 2.9 x speedup without sacrificing accuracy. Our code is available at: https://github.com/sharc-lab/Residual-INR.

Azita Nouri, Philip E. Davis, Pradeep Subedi et al.

Graph embedding techniques have attracted growing interest since they convert the graph data into continuous and low-dimensional space. Effective graph analytic provides users a deeper understanding of what is behind the data and thus can benefit a variety of machine learning tasks. With the current scale of real-world applications, most graph analytic methods suffer high computation and space costs. These methods and systems can process a network with thousands to a few million nodes. However, scaling to large-scale networks remains a challenge. The complexity of training graph embedding system requires the use of existing accelerators such as GPU. In this paper, we introduce a hybrid CPU-GPU framework that addresses the challenges of learning embedding of large-scale graphs. The performance of our method is compared qualitatively and quantitatively with the existing embedding systems on common benchmarks. We also show that our system can scale training to datasets with an order of magnitude greater than a single machine's total memory capacity. The effectiveness of the learned embedding is evaluated within multiple downstream applications. The experimental results indicate the effectiveness of the learned embedding in terms of performance and accuracy.

Jim Samuel, Margaret Brennan-Tonetta, Yana Samuel et al.

There has been an increasing interest in and growing need for high performance computing (HPC), popularly known as supercomputing, in domains such as textual analytics, business domains analytics, forecasting and natural language processing (NLP), in addition to the relatively mature supercomputing domains of quantum physics and biology. HPC has been widely used in computer science (CS) and other traditionally computation intensive disciplines, but has remained largely siloed away from the vast array of social, behavioral, business and economics disciplines. However, with ubiquitous big data, there is a compelling need to make HPC technologically and economically accessible, easy to use, and operationally democratized. Therefore, this research focuses on making two key contributions, the first is the articulation of strategies based on availability, accessibility and usability for the demystification and democratization of HPC, based on an analytical review of Caliburn, a notable supercomputer at its inception. The second contribution is a set of principles for HPC adoption based on an experiential narrative of HPC usage for textual analytics and NLP of social media data from a first time user perspective. Both, the HPC usage process and the output of the early stage analytics are summarized. This research study synthesizes expert input on HPC democratization strategies, and chronicles the challenges and opportunities from a multidisciplinary perspective, of a case of rapid adoption of supercomputing for textual analytics and NLP. Deductive logic is used to identify strategies which can lead to efficacious engagement, adoption, production and sustained usage for research, teaching, application and innovation by researchers, faculty, professionals and students across a broad range of disciplines.