A Genai-Driven Risk Prediction Framework For Program Management In Embedded Media Platforms

Machine learning has surged in popularity in recent years, with deep neural networks (DNNs) becoming a cornerstone of applications such as autonomous driving, augmented reality, and natural language processing, owing to their remarkable accuracy and scalability. As the demand for greater portability and flexibility grows, there is a strong push to optimize these applications for resource-constrained edge devices like smartphones and IoT devices.However, deploying DNN models on such devices presents significant challenges, primarily due to the large size of these models, which is a key contributor to their high performance. For instance, the DeepSpeech 2 model outperforms its predecessor, DeepSpeech 1, but its size is eight times larger, encompassing approximately 68 million parameters. Similarly, other networks designed for various tasks, including convolutional neural networks like VGG and transformer-based networks like BERT and GPT, also exhibit substantial sizes. For example, the GPT-3 model, a predecessor to ChatGPT, consists of an astonishing 175 billion parameters. Consequently, fitting DNN models onto resource-limited edge devices is a formidable challenge. My thesis addresses these challenges with a three-part approach: * High-Performance DNN Accelerators: I will develop DNN accelerators that excel in high performance, low latency, and energy efficiency through an algorithm and hardware co-design across diverse FPGA architectures (Chapter 2). * Real-Time DNN Execution: I aim to achieve exceptional accuracy and real-time execution for various large-scale DNNs on off-the-shelf mobile devices, IoT devices, and other platforms (Chapter 3). * CompVQC Framework: I will explore a systematic framework, CompVQC, to reduce the circuit length of quantum neural networks (QNNs) through model compression, which can achieve up to a 20% improvement in accuracy. This reveals that compilation-aware compression can enhance the robustness of QNNs on near-term noisy quantum devices (Chapter 4). The field of AI has recently undergone a significant transformation, ushering in the era of generative AI, where models like ChatGPT and Stable Diffusion create realistic and creative content for a wide range of applications. As generative AI becomes more prevalent, it is crucial to adapt algorithm-hardware co-design strategies to meet the specific needs of these models, particularly on edge devices. In the final chapter (Chapter 5), I will introduce my ongoing and future work focused on activating generative AI on the edge.--Author's abstract

<p style='text-indent:20px;'>In this paper, we study linear transport model by adopting <i>deep learning method</i>, in particular deep neural network (DNN) approach. While the interest of using DNN to study partial differential equations is arising, here we adapt it to study kinetic models, in particular the linear transport model. Moreover, theoretical analysis on the convergence of neural network and its approximated solution towards analytic solution is shown. We demonstrate the accuracy and effectiveness of the proposed DNN method in numerical experiments.</p>

This chapter discusses the opportunities and challenges posed by the use of generative AI in the Sharia auditing profession. Generative AI (GAI) technologies have the potential to transform the operations of the financial and auditing departments completely. In a way, they perform self-importance roles such as building insights and automating challenging tasks while enhancing decision-making in these sectors. It is evident that through the use of GAI, the power of deep learning and neural networks can be harnessed to improve business processes and monitor outliers, which are critical for achieving both compliance and transparency in finance. However, we must highlight certain drawbacks that come along when employing GAI for Sharia-compliant auditing. According to Islamic law, Sharia compliance is based on Shariah rules and laws, so the audit has to meet certain requirements. Hence, auditing has to obey certain ethical and legal aspects.Does GAI integration into Sharia-compliant audits carry a concern of operating outside the technological, ethical, and Islamic principles? Yes, it has been observed that GAI systems depend on algorithmic geography, data tenderness, and the rules of regulation for their use. Besides, the GAI uptake exposes the increasing need for Sharia data and gaps in the interpretability of existing AI models. These are some of the reasons why GAI for Sharia auditors is attractive, even as there are challenges like automation of compliance and fraud detection, and a gap in operational definitions.This chapter explores the implication of GAI on Sharia auditors in terms of contemporary practice impact, challenges like ethics and technical limitations, and new opportunities for efficiency and compliance improvements. This discussion embeds insights from recent research, including works like Rabbani et al. on the necessity for ethical frameworks relating to AI in Islamic finance and Brown et al.’s findings about advanced language models in terms of their competency for actionable insight performance. Based on these aspects, this chapter will seek an all-rounded appreciation of the challenges and opportunities generative AI does or will pose.

Tourism is one of the most important sectors of the global economy, generating billions of euros in revenue and supporting millions of jobs. However, the tourism industry also faces many challenges, such as changing consumer preferences, environmental impacts and competitive pressures. In this paper, we explore how specialized and generative AI, a branch of artificial intelligence that can create novel content, such as text, images, or audio, can support hotels and other tourist accommodation companies in enhancing their products and services, improving their marketing and communication strategies and optimizing their operations and management. We review the existing literature on AI and its applications in tourism and propose potential use cases of generative AI for tourist companies. We also present some examples of generative AI tools that can be used for these purposes and discuss their benefits and limitations. Finally, we offer some conclusions and future implications for research and practice in this emerging field. Theoretical implications of the present paper suggest that the incorporation of generative AI in the tourism sector can significantly advance the current understanding of AI applications in service industries. By leveraging machine learning techniques, such as deep neural networks and generative adversarial networks, the tourism industry can create novel content and optimize various operations. This study enriches existing literature by identifying new use cases for generative AI, thus providing a foundation for further academic inquiry. Practically, our findings reveal several actionable insights for tourism companies. Generative AI tools can be harnessed to personalize customer experiences, enhance marketing efforts, and streamline operational management.

Designing new hit series of JAK3 inhibitors using generative AI, reinforcement learning, and molecular dynamics.

Heating, Ventilation, and Air Conditioning (HVAC) systems are major energy consumers in buildings, challenging the balance between efficiency and occupant comfort. While prior research explored generative AI for HVAC control in simulations, real-world validation remained scarce. This study addresses this gap by designing, deploying, and evaluating “Office-in-the-Loop,” a novel cyber-physical system leveraging generative AI within an operational office setting. Capitalizing on multimodal foundation models and Agentic AI, our system integrates real-time environmental sensor data (temperature, occupancy, etc.), occupants’ subjective thermal comfort feedback, and historical context as input prompts for the generative AI to dynamically predict optimal HVAC temperature setpoints. Extensive real-world experiments demonstrate significant energy savings (up to 47.92%) while simultaneously improving comfort (up to 26.36%) compared to baseline operation. Regression analysis confirmed the robustness of our approach against confounding variables like outdoor conditions and occupancy levels. Furthermore, we introduce Data-Driven Reasoning using Agentic AI, finding that prompting the AI for data-grounded rationales significantly enhances prediction stability and enables the inference of system dynamics and cost functions, bypassing the need for traditional reinforcement learning paradigms. This work bridges simulation and reality, showcasing generative AI’s potential for efficient, comfortable building environments and indicating future scalability to large systems like data centers.

Learning ability is the basic characteristic of human intelligence. The July 1, 2005 issue of Science published a list of 125 important questions in science. Among them, the question 94 “What are the limits of learning by machines?”. The annotation “Computers can already beat the world’s best chess players, and they have a wealth of information on the Web to draw on. But abstract reasoning is still beyond any machine”. In recent artificial intelligence has made great progresses. In 1997, the rise of the man-machine war, IBM Supercomputer Deep Blue defeated the chess master Garry Kasparov. On February 14, 2011, IBM’s Watson supercomputer won a practice round against Jeopardy champions Ken Jennings and Brad Rutter. In March 2016, Google DeepMind’s AlphaGo sealed a 4-1 victory over a South Korean Go grandmaster Lee Se-dol. This paper focuses on the machine learning methods of AlphaGo, including reinforcement learning, deep learning, deep reinforcement learning, analysis of the existing problems and the latest research progress. Deep reinforcement learning is the combination of deep learning and reinforcement learning, which can realize the learning algorithm from the perception to action. Simply said, this is the same as human behavior, input sensing information such as vision, and then, direct output action through the deep neural network. Deep reinforcement learning has the potential to learn a variety of skills for the robot to achieve full autonomy. Even though reinforcement learning is practiced successfully, but feature states need to manually set, for complex scene is a difficult thing, especially easy to cause the dimension disaster, and expression is not good. In 2010, Sascha Lange and Martin Riedmiller proposed deep auto-encoder neural networks in reinforcement learning to extract feature, which is used to control the visual correlation. In 2013, DeepMind proposed deep Q-network (DQN) in NIPS 2013, using convolution neural network to extract features, and then applied in reinforcement learning. They continue to improve and published an improved version of DQN on Nature in 2015, which has aroused widespread concern. In order to break through the limits of learning by machines, cognitive machine learning is proposed, which is the combination of machine learning and brain cognition, so that the machine intelligence is constantly evolving, and gradually reaches the human level of artificial intelligence. A cognitive model entitled Consciousness And Memory (CAM) is proposed by author, which consists of memory, consciousness, high-level cognitive functions, perception and motor. High-level cognitive functions of the brain include learning, language, thinking, decision making, emotion, and so on. Learning is a course to accept the stimulus through the nervous system and obtain new behavior, habits and accumulation experience. According to the current research progress of brain science and cognitive science, cognitive machine learning may be interested in learning emergence, procedural memory knowledge learning, learning evolution and so on. For intelligence, so-called evolution is refers to the learning of learning and the structure also follows the change. It is important to record the learning result by structure changing and improve the learning method.

Generative AI, a subset of artificial intelligence, is dedicated to the development of models that are capable of producing new content, including text, imagery, audio, and entire virtual worlds. Generative AI is capable of replicating human creativity and generating outputs that were previously considered to be exclusively the domain of human intelligence by utilizing sophisticated machine learning techniques, including deep learning and neural networks. AI models in the music industry are capable of composing music in a variety of genres, providing musicians and producers with new tools. Generative AI has the potential to stimulate innovation and creativity in a variety of disciplines as technology continues to develop. This research article delves deep into the various aspects of Generative AI namely history and core concepts, privacy and security, types of Generative AI models and their applications, NLP models and finally on Responsible AI.

ABSTRACTRecently, the popularity of deep artificial neural networks has increased considerably. Generally, the method used in the training of these structures is simple gradient descent. However, training a deep structure with simple gradient descent can take quite a long time. Some additional approaches have been utilized to solve this problem. One of these techniques is the momentum that accelerates gradient descent learning. Momentum techniques can be used for supervised learning as well as for reinforcement learning. However, its efficiency may vary due to the dissimilarities in two learning processes. While the expected values of inputs are clearly known in supervised learning, it may take long-running iterations to reach the exact expected values of the states in reinforcement learning. In an online learning approach, a deep neural network should not memorize and continue to converge with the more precise values that exist over time during these iterations. For this reason, it is necessary to use a momentum technique that both adapt to reinforcement learning and accelerate the learning process. In this paper, the performance of different momentum techniques is compared with the Othello game benchmark. Test results show that the Nesterov momentum technique provided a more effective generalization with an online reinforcement learning approach.

BackgroundEpigenetic modification has an effect on gene expression under the environmental alteration, but it does not change corresponding genome sequence. DNA methylation (DNAm) is one of the important epigenetic mechanisms. DNAm variations could be used as epigenetic markers to predict and account for the change of many human phenotypic traits, such as cancer, diabetes, and high blood pressure. In this study, we built deep neural network (DNN) regression models to account for interindividual variation in triglyceride concentrations measured at different visits of peripheral blood samples using epigenome-wide DNAm profiles.ResultsWe used epigenome-wide DNAm profiles of before and after medication interventions (called pretreatment and posttreatment, respectively) to predict triglyceride concentrations for peripheral blood draws at visit 2 (using pretreatment data) and at visit 4 (using both pretreatment and posttreatment data). Our experimental results showed that DNN models can predict triglyceride concentrations for blood draws at visit 4 using pretreatment and posttreatment DNAm data more accurately than for blood draws at visit 2 using pretreatment DNAm data. Furthermore, we got the best prediction results when we used pretreatment DNAm data to predict triglyceride concentrations for blood draws at visit 4, which suggests a long-term epigenetic effect on phenotypic traits. We compared the prediction performances of our proposed DNN models with that of support vector machine (SVM). This comparison showed that our DNN models achieved better prediction performance than did SVM.ConclusionsWe demonstrated the superiority of our proposed DNN models over the SVM model for predicting triglyceride concentrations. This study also suggests that the DNN approach has advantages over other traditional machine-learning methods to model high-dimensional epigenome-wide DNAm data and other genomic data.

The advent of generative AI has opened up more avenues for mental health challenges among youth worldwide. This work represents a comprehensive survey of the recent progress of generative AI in youth mental healthcare concerning their impact, enabling technologies, challenges, and future directions. Some of the key technologies that form the bedrock for many applications involve NLP, transformer models, GAN, VAE, reinforcement learning, and Affective AI in the development of AI-powered chatbots, virtual therapists, personalized interventions, and early detection. We assess the role of open-source technologies such as orchestration frameworks, vector databases, large and small language models, front-end tools, and other enablers that empower RAG systems to elevate the capability for contextual relevance and effectiveness in Generation AI solutions. We present the positive impacts of generative AI: enhanced accessibility, increased engagement, and decreased cost, while also discussing critical concerns on ethical issues, data privacy, bias, and informed consent. We also probe barriers to mainstreaming, including technical limitations, policy gaps, and social resistance, and presented a critical review of pathways toward surmounting the difficulties. In the final portion of the study, we make conclusions and recommendations with regard to strategic future research and policy development recommendations that will ensure that generative AI is safe, ethical, and effective to deploy within youth mental healthcare for improved mental health outcomes among young populations globally.

<p dir="ltr">Objective: We aim to develop a multimodal model to predict chronic kidney disease (CKD) in patients with type 2 diabetes mellitus (T2DM), given the limited research on this integrative approach.</p><p dir="ltr">Research Design and Methods: We obtained multimodal datasets from Kyung Hee University Medical Center (South Korea; n=7028; discovery cohort) for training and internal validation and UK Biobank (n=1544; validation cohort) for external validation. CKD was defined based on the international classification of disease-9/10 codes and/or estimated glomerular filtration rate (eGFR) ≤60 mL/min/1.73m². We ensembled various deep learning models and interpreted their predictions using explainable artificial intelligence (AI) methods, including shapley additive explanations (SHAP) and gradient-weighted class activation mapping (Grad-CAM). Subsequently, we investigated the potential association between the model probability and vascular complications.</p><p dir="ltr">Results: The multimodal model, which ensembles visual geometry group 16 (VGG16) and deep neural network (DNN), presented high performance in predicting CKD, with area under the receiver operating characteristic curve (AUROC) of 0.880 (95% CI, 0.806–0.954) in the discovery cohort and 0.722 in the validation cohort. SHAP and Grad-CAM highlighted key predictors, including eGFR and optic disc, respectively. The model probability was associated with an increased risk of macrovascular complications (T1: adjusted hazard ratio, 1.42 [95% CI, 1.06–1.90]; T2: 1.59 [1.17–2.16]; and T3: 1.64 [1.20–2.26]) and microvascular complications (T3: 1.30 [1.02–1.67]).</p><p dir="ltr">Conclusions: Our multimodal AI model integrates fundus images and clinical data from binational cohorts to predict the risk of new-onset CKD within five years and associated vascular complications in patients with T2DM.</p>

Generative AI (GenAI), exemplified by Large Language Models (LLMs), such as OpenAI’s ChatGPT, is revolutionizing various fields. Central to this transformation is Data Center Networking (DCN), which not only provides the infrastructure support for GenAI operation, but also provisions GenAI services to users. Hence, this article explores the interplay between GenAI and DCNs, analyzing their symbiotic relationship and mutual advances. We begin by reviewing the current challenges of DCNs and GenAI-based solutions, such as data augmentation, process automation, and domain transfer. We then discuss the distinctive characteristics of GenAI workloads on DCNs, gaining insights that catalyze the evolution of DCNs to more effectively support GenAI. Moreover, to illustrate the seamless integration of GenAI with DCNs, we present a case study on GenAI-empowered DCN digital twins. Specifically, we employ an LLM equipped with retrieval augmented generation to formulate optimization problems for DCNs (e.g., resource allocation and routing) and adopt diffusion-deep reinforcement learning to solve optimization. The experimental results on a representative DCN optimization problem, i.e., knowledge placement, demonstrate the validity and efficiency of our proposals. We anticipate that this article can promote further research to enhance the virtuous interaction between GenAI and DCNs.

The paramount challenge in precision oncology lies in further improving quality of life and response rates for individual patients. Efforts toward these goals are steadily expanding the scope of clinical implementation, despite ongoing challenges such as standardization, cost-effectiveness, and data harmonization. Building upon this maturing foundation, generative AI—which has evolved dramatically in recent years—is particularly valuable at this stage of advancing efficiency and adoption as an auxiliary technology linking literature, guidelines, trial protocols, and patient data. Specifically, through mutation interpretation, trial eligibility matching, and tumor board support, it is expected to contribute to advancing standardization, improving cost-effectiveness, accelerating data harmonization, and further accelerating human-centered decision-making. Accordingly, this review surveys the development history of generative AI and its current healthcare applications, organizing its implementation potential for precision oncology along three axes: (1) generative AI–based interpretation of genetic mutations and estimation of their pathological significance; (2) generative AI–driven verification of clinical trial eligibility; and (3) multimodal foundation models for imaging and pathology that compute “tumor phenotypes” using real-world data, contributing to report drafting and molecular surrogate estimation. In response, we propose a strategy centered on retrieval-augmented generation (RAG) and human-in-the-loop (HITL) workflows, encompassing data preparation based on OMOP, mCODE, and FHIR; multicenter prospective evaluation; auditable logs and governance aligned with Good Manufacturing Practice (GMP) and the EU AI Act; and a synthetic data strategy including differential privacy. Ultimately, this approach validates value through real-world outcomes and charts a path toward “learning oncology,” accelerating patient-centered decision-making and clinical trial development.

Deep neural networks (DNNs) have achieved unprecedented success across many scientific and engineering fields in the last decades. Despite its empirical success, unfortunately, recent studies have shown that there are various failure modes and blindspots in DNN models which may result in unexpected serious failures and potential harms, e.g. the existence of adversarial examples and small perturbations. This is not acceptable especially for safety critical and high stakes applications in the real-world, including healthcare, self-driving cars, aircraft control systems, hiring and malware detection protocols. Moreover, it has been challenging to understand why and when DNNs will fail due to their complicated structures and black-box behaviors. Lacking interpretability is one critical issue that may seriously hinder the deployment of DNNs in high-stake applications, which need interpretability to trust the prediction, to understand potential failures, and to be able to mitigate harms and eliminate biases in the model. To make DNNs trustworthy and reliable for deployment, it is necessary and urgent to develop methods and tools that can (i) quantify and improve their robustness against adversarial and natural perturbations, and (ii) understand their underlying behaviors and further correct errors to prevent injuries and damages. These are the important first steps to enable Trustworthy AI and Trustworthy Machine Learning. In this talk, I will survey a series of research efforts in my lab contributed to tackling the grand challenges in (i) and (ii). In the first part of my talk, I will overview our research effort in Robust Machine Learning since 2017, where we have proposed the first attack-agnostic robustness evaluation metric, the first efficient robustness certification algorithms for various types of perturbations, and efficient robust learning algorithms across supervised learning to deep reinforcement learning. In the second part of my talk, I will survey a series of exciting results in my lab on accelerating interpretable machine learning and explainable AI. Specifically, I will show how we could bring interpretability into deep learning by leveraging recent advances in multi-modal models. I'll present recent works in our group on automatically dissecting neural networks with open vocabulary concepts, designing interpretable neural networks without concept labels, and briefly overview our recent efforts on demystifying black-box DNN training process, automated neuron explanations for Large Language Models and the first robustness evaluation of a family of neuron-level interpretation techniques.