Electrical and Computer Engineering Commons^™

Cooperative Deep Q -Learning Framework For Environments Providing Image Feedback, Krishnan Raghavan, Vignesh Narayanan, Sarangapani Jagannathan

Publications

In this article, we address two key challenges in deep reinforcement learning (DRL) setting, sample inefficiency, and slow learning, with a dual-neural network (NN)-driven learning approach. In the proposed approach, we use two deep NNs with independent initialization to robustly approximate the action-value function in the presence of image inputs. In particular, we develop a temporal difference (TD) error-driven learning (EDL) approach, where we introduce a set of linear transformations of the TD error to directly update the parameters of each layer in the deep NN. We demonstrate theoretically that the cost minimized by the EDL regime is an approximation …

A Semantic Web Approach To Fault Tolerant Autonomous Manufacturing, Fadi El Kalach, Ruwan Wickramarachchi, Ramy Harik, Amit Sheth

Publications

The next phase of manufacturing is centered on making the switch from traditional automated to autonomous systems. Future factories are required to be agile, allowing for more customized production, and resistance to disturbances. Such production lines would be able to reallocate resources as needed and minimize downtime while keeping up with market demands. These systems must be capable of complex decision-making based on parameters such as machine status, sensory/IoT data, and inspection results. Current manufacturing lines lack this complex capability and instead focus on low-level decision-making on the machine level without utilizing the generated data to its full extent. This …

Knowledge Graph Guided Semantic Evaluation Of Language Models For User Trust, Kaushik Roy, Tarun Garg, Vedant Palit, Yuxin Zi, Vignesh Narayanan, Amit Sheth

Publications

A fundamental question in natural language processing is - what kind of language structure and semantics is the language model capturing? Graph formats such as knowledge graphs are easy to evaluate as they explicitly express language semantics and structure. This study evaluates the semantics encoded in the self-attention transformers by leveraging explicit knowledge graph structures. We propose novel metrics to measure the reconstruction error when providing graph path sequences from a knowledge graph and trying to reproduce/reconstruct the same from the outputs of the self-attention transformer models. The opacity of language models has an immense bearing on societal issues of …

Acm Web Conference 2023, Usha Lokala, Kaushik Roy, Utkarshani Jaimini, Amit Sheth

Publications

Improving the performance and explanations of ML algorithms is a priority for adoption by humans in the real world. In critical domains such as healthcare, such technology has significant potential to reduce the burden on humans and considerably reduce manual assessments by providing quality assistance at scale. In today’s data-driven world, artificial intelligence (AI) systems are still experiencing issues with bias, explainability, and human-like reasoning and interpretability. Causal AI is the technique that can reason and make human-like choices making it possible to go beyond narrow Machine learning-based techniques and can be integrated into human decision-making. It also offers intrinsic …

The Troubling Emergence Of Hallucination In Large Language Models--An Extensive Definition, Quantification, And Prescriptive Remediations, Vipula Rawte, Swagata Chakraborty, Agnibh Pathak, Anubhav Sarkar, S.M Towhidul Islam Tonmoy, Aman Chadha, Amit Sheth, Amitava Das

Publications

The recent advancements in Large Language Models (LLMs) have garnered widespread acclaim for their remarkable emerging capabilities. However, the issue of hallucination has parallelly emerged as a by-product, posing significant concerns. While some recent endeavors have been made to identify and mitigate different types of hallucination, there has been a limited emphasis on the nuanced categorization of hallucination and associated mitigation methods. To address this gap, we offer a finegrained discourse on profiling hallucination based on its degree, orientation, and category, along with offering strategies for alleviation. As such, we define two overarching orientations of hallucination: (i) factual mirage (FM) …

Development Of Holographic Recording Material For Fabrication Of Solar Concentrators For Micro Satellites, Brian Rogers

Doctoral

Holographic solar concentration has been a topic of increasing interest over recent years because of the lightweight and physically compact nature of holograms, which gives them an advantage over other, conventional concentration methods. As holographic recording materials develop further, their characteristics prove to meet the demands required of them to make more effective holographic solar concentrators. The development of photovoltaics in tandem with holographic materials has led to a host of solar concentrating device designs that take advantage of the unique properties of diffractive optics. Compared to refractive optics, holographic devices are more lightweight and compact. This makes holographic devices …

Investigating The Effects Of Network Dynamics On Quality Of Delivery Prediction And Monitoring For Video Delivery Networks, Obinna C. Izima

Doctoral

Video streaming over the Internet requires an optimized delivery system given the advances in network architecture, for example, Software Defined Networks. Machine Learning (ML) models have been deployed in an attempt to predict the quality of the video streams. Some of these efforts have considered the prediction of Quality of Delivery (QoD) metrics of the video stream in an effort to measure the quality of the video stream from the network perspective. In most cases, these models have either treated the ML algorithms as black-boxes or failed to capture the network dynamics of the associated video streams.

This PhD investigates …

Comprehensive Fault Diagnostics And Performance Forecasting Of Wind Turbines Through Condition Monitoring Solutions, Shuo Zhang

Doctoral

The operation and maintenance (O&M) issues of wind turbines (WTs) are challenging because unplanned maintenance, caused by sudden component failures, can bring about durable downtimes and significant revenue losses. It is important to carry out effective fault diagnostics and prognostics schemes under the rapid development of wind power generation. Hence, Condition Monitoring (CM) solutions focus on measurements and detections of high-risk WT components, which could result in high failure rates and long downtimes. Machine Learning (ML) models have been commonly applied in CM to allocate imminent indications of failure or degradation for curtailing the O&M costs of WTs. By way …