Digital Commons Network^™

Optimizing Coupon Distribution By On-Demand Allocation Of Voucher Codes, Ankit Aggarwal, Arpit Kumar Jaiswal, Woohyun Kim

Defensive Publications Series

Voucher codes attached to digital scratch coupons are a popular way for merchants to present offers to potential customers. Such coupons are distributed through digital payment apps, online shopping platforms, etc. after a successful payment by a user. A substantial fraction of coupons may not get interacted with by the user. Even when the coupon is activated, many users may not view or use the voucher code. This causes wasteful effort for marketers that run promotional campaigns linked to coupons. This disclosure describes techniques to assign and distribute voucher codes only after a user seeks the code after scratching a …

Enhanced Health And Memory Augmentation System: An Integrated Approach To Age-Adaptive Health Monitoring And Cognitive Support, Ramkumar Bharathan

Defensive Publications Series

This defensive publication discloses a novel system that integrates memory augmentation with comprehensive health monitoring and personalized concierge services for users across all age demographics. The Enhanced Memory and Health Augmentation System (EMHAS) utilizes artificial intelligence to create a secure, privacy-preserving platform that continuously collects, analyzes, and correlates multimodal data including health metrics, behavioral patterns, and personal memories. Key innovations include: (1) a Selective Memory Capture Architecture with age-adaptive interfaces; (2) a Health Integration and Monitoring Engine for continuous biometric tracking; (3) a Federated Personalization Engine employing on-device processing for privacy protection; (4) a Contextual Memory Retrieval System providing anticipatory …

Execution Of Commands Using An Ai Engine, Nixon Aw, Bryson Lee, Yufeng Zhuang, Anita Wan

Defensive Publications Series

One embodiment includes a method comprising: training an artificial intelligence (AI) engine with successfully executed prior action data; receiving, by the AI engine, a request to perform an action; identifying, by the AI engine, one or more attributes to execute the action; obtaining, by the AI engine, a confirmation of the one or more attributes; identifying, by the AI engine, a plurality of APIs to perform the action as defined by the one or more attributes; identifying, by the AI engine, a sequence of execution for the plurality of APIs to perform the action; and executing, by the AI engine, …

Failure Characterization Of Aluminum Resistance Spotwelds Using Ls-Dyna & Developing Machine Learning Model To Predict Spotweld Failure Parameters And Test Results, Akshay Arthur Kulkarni, Joao Felipe Carvalho De Moraes, Aaditya Lakshmanan, Sazol Kumar Das, Lily Ann Slabbert, Ashwin Yeagappan

Defensive Publications Series

[0001] This project developed methodology to create failure cards for aluminum resistance spotwelds using LS-DYNA, with a project objective to achieve reasonable accuracy (>90%). Rigorous testing and simulation activities were conducted while developing the methodology. The methodology includes mathematically fitting a failure model to match the test data efforts and deviation from the test data. Weld development and the testing process included weld schedule development through robustness testing and microhardness testing to study heat affected zones around the weld nugget properties. Welded coupons were cross-sectioned to analyze for cracks and porosity in the weld nugget. Eventually, lap shear, coach …

Mastering Technology Audit Quality Assurance: A Framework For Auditors, Ramkumar Bharathan

Defensive Publications Series

Ramkumar's document presents a comprehensive framework (COBIT, NIST, FFIEC CAT, FFIEC IT Handbook) for establishing, implementing, and maturing a robust Technology Quality Assurance Review (QAR) program within internal audit, emphasizing its strategic imperative in today's complex digital landscape. Driven by escalating technology risks (cybersecurity, cloud, AI), evolving regulatory demands (FFIEC, COBIT, NIST CSF, GLBA, SOX, NYDFS, State Privacy Laws), and the need for reliable assurance, the framework details a systematic approach grounded in IIA Global Internal Audit Standards (specifically the QAIP) and integrated with key governance and control frameworks like COBIT and NIST CSF. It outlines core QAR principles, a …

Context-Aware Dynamic Diagram Generation For Enhanced Whiteboarding Sessions In An Online Meeting, Ram Mohan R, Rajarshee Dhar, Deepesh Arora

Defensive Publications Series

Proposed herein are techniques that provide for the ability to derive context for an online meeting from various contextual sources. During operation, the system proposed herein can predict a user's diagramming intent and then dynamically populate a whiteboarding palette with relevant diagrams based on the context of the online meeting.

Protecting Security And Privacy In Mobile Ai Ecosystems, Zongyong (Vincent) Chen

Defensive Publications Series

This publication describes systems and methods for protecting security and privacy in mobile artificial intelligence (AI) ecosystems. The use of protected virtual machines (pVMs) serves to isolate the AI components of the ecosystem, ensuring both privacy and security for end-user data. By running the AI on a separate operating system (OS), even a compromise of the main OS does not threaten the integrity of the data and output of the AI. In addition, the AI can be implemented on separate hardware than that of the main OS, such as separate processor(s), memory, etc., making user data inaccessible to malicious actors.

Wifi Access Point Placement And Configuration Using Reinforcement Learning Agents, N/A

Defensive Publications Series

Reliable WiFi coverage hinges on the strategic placement and accurate configuration of access points (APs). Traditional, manual AP placements are time consuming, costly, and suboptimal. This disclosure describes techniques that leverage Q-learning to determine optimal access point (AP) placements and configurations in an indoor environment. A reinforcement learning (RL) agent is trained using a digital representation of the indoor environment and a reward function that maximizes network coverage while minimizing the number of APs. The digital representation of the indoor environment includes floorplans and coverage maps. The trained RL agent can be deployed to optimize the placement and configuration of …

Wifi Access Point Placement And Configuration Using A* Search, N/A

Defensive Publications Series

Reliable WiFi coverage hinges on the strategic placement and accurate configuration of access points (APs). Traditional, manual AP placements are time consuming, costly, and suboptimal. This disclosure describes techniques that leverage the A* search algorithm to determine optimal AP placements/configurations. Given a set of AP placements/configurations and a metric that captures the difference between the network coverage of a particular AP placement and a target network coverage, an AP placement/configuration is found that optimizes WiFi coverage while minimizing the number of APs and their configuration complexities. The techniques advantageously discretize the search space by leveraging the facts that only a …

Saving Idle Power And Engineering Hours By Powering Off/On Idle Ai Accelerators, N/A

Defensive Publications Series

The technology described in this paper relates to an artificial intelligence (AI) accelerator chip's application-specific integrated circuits (ASICs)’ power down for idle power reduction, and low-latency peripheral component interconnect express (PCIe) relink made on AI accelerator chips. Complete turning on or off of the AI accelerator chips may be accomplished by leveraging various software and/or hardware components in the AI accelerator systems. The specialized software and/or hardware may be configured to turn off the entirety or only certain components of each AI accelerator chip, such as application-specific integrated circuits (ASICs) embedded in the AI accelerator chip, and remove individual PCIe …

Scalable Real-Time Bias Detection And Mitigation In Ai-Based Application Evaluation, Puneet Mahajan

Defensive Publications Series

Employing artificial intelligence (AI) for evaluating AI applications poses a number of challenges, including the detection and mitigation of various biases. This disclosure describes a multi-faceted approach for dynamically identifying and mitigating biases in AI-based evaluation. Potential biases can be detected by analyzing inconsistencies and discrepancies between the outputs of the primary and auxiliary evaluator models via an ensemble learning framework, optionally assisted by human evaluators. Specific adversarial examples can be generated and used to identify hidden biases in the primary AI evaluator. Detected biases can be mitigated by employing reinforcement learning to make continuous real-time adjustments to counteract the …

Aggn: Adaptive Granularity Graph Networks For Heterophilic Environments

Defensive Publications Series

Graph neural networks (GNNs) have shown significant success in learning graph representations. However, recent studies reveal that GNNs often fail to outperform simple MLPs on heterophilous graph tasks, where connected nodes may differ in features or labels, challenging the homophily assumption. Existing methods addressing this issue often overlook the importance of information granularity and rarely consider implicit relationships between distant nodes. To overcome these limitations, we propose the Adaptive Granularity Graph Network (AGGN), a novel GNN model specifically designed for heterophilous graphs. AGGN enhances node embeddings by aggregating multi-view information at various granularity levels and incorporating implicit data from distant, …

Halo: Heterophily-Aware Label-Free Ordering For Unsupervised Graph Fraud Detection

Defensive Publications Series

Graph fraud detection (GFD) has rapidly advanced in protecting online services by identifying malicious fraudsters. Recent supervised GFD research highlights that heterophilic connections between fraudsters and users can greatly impact detection performance, since fraudsters tend to camouflage themselves by building more connections to benign users. Despite the promising performance of supervised GFD methods, the reliance on labels limits their applications to unsupervised scenarios; Additionally, accurately capturing complex and diverse heterophily patterns without labels poses a further challenge. To fill the gap, we propose a Heterophily-guided Unsupervised Graph fraud dEtection approach (HUGE) for unsupervised GFD, which contains two essential components: a …

Learning Resilient Graph Structures In Heterophilic Settings

Defensive Publications Series

Graphs provide a fundamental way to model relationships between entities and are central to numerous machine learning tasks. Standard graph-based methods often assume the provided graph structure is both accurate and complete. However, real-world graphs frequently suffer from noise and sparsity, negatively impacting downstream tasks like node classification and clustering. While graph representation learning has advanced significantly, many methods implicitly assume graph homophily (connections predominantly between nodes of the same class), struggling when faced with heterophily (connections predominantly between different classes). This paper introduces a novel method, Resilient Graph Learning for Heterophily (RGLH), designed to learn high-quality graph structures directly …

Quantum-Coherent Graphene Via Entangled Plasma Deposition (Coherium™) Room Temp Superconductor, Peter Branton

Defensive Publications Series

Twisted bilayer graphene (TBG) at a magic angle (~1.1°) has demonstrated superconductivity below ~1.7 K due to emergent flat-band behaviour. However, its mechanical assembly limitations, thermal fragility, and scalability issues hinder real-world applications. The Coherium™ process addresses these limitations by synthesizing entangled carbon plasma into a structurally coherent graphene lattice with integrated AI stabilization, targeting room-temperature Superconducting properties.

Efficient Partitioning For Devices With Multiple User Modes, Harshad Dhabu, Abhijit Adsule

Defensive Publications Series

This document describes techniques that enable a computing device (e.g., a smart watch, ring, glasses, etc.) to more efficiently transition between multiple user modes, such as an adult mode and a child mode. Using a single system image including applications installed on the computing device for every user mode may lead to system image bloat, diminished performance and power inefficiency. Alternatively, creating separate stock keeping units (SKUs) for each user mode may introduce manufacturing and inventory complexities. Instead, the device may divide the device's storage and the applications corresponding to each user mode into three partitions. The first partition may …

Machine Learning System And Method For Predicting Risk Of Drilling Component Failure, Baker Hughes Company

Defensive Publications Series

The invention addresses the problem of drilling tool failure risk assessment in the oil and gas industry. Tool failure in drilling operations can lead to significant financial losses, downtime, and safety risks. The invention aims to provide a solution for predicting and preventing these failures, which is crucial for maintaining safe and efficient drilling operations.

The invention proposes a solution by combining data-driven techniques with expert knowledge. It begins with data collection, including historical data on tool performance, and involves data cleaning and pre-processing. Relevant features are identified, and a predictive model is developed using machine learning techniques. The model …

Quantum Smart Entangled Fusion Reactor, Peter Branton

Defensive Publications Series

The invention relates to a compact, high-efficiency energy generation system employing

quantum-entangled hydrogen isotope fusion within nanoscale confinement structures. It

comprises a multiplicity of individually controlled fusion traps formed at nanometer scale, each

configured to receive pre-entangled pairs of hydrogen isotopes such as protium, deuterium, or

tritium. Each trap is stimulated by attosecond-timed electromagnetic resonance pulses to induce

fusion via quantum tunneling. Fusion events are monitored and optimised through an integrated

artificial intelligence (AI) control layer that ensures only coherence-validated, phase-aligned

pairs are activated. The system includes capacitive and inductive energy reclamation

components, and operates in inert gas environments without …

Ts-Ratgnn: Two-Stage Spectral Graph Neural Network With Explicitly Optimized Rational Filters

Defensive Publications Series

Approximation-based spectral graph neural networks, which construct graph filters via function approximation, have demonstrated significant success in graph learning tasks. While existing methods predominantly rely on polynomial approximations for filter construction, rational approximation, a potentially superior alternative, remains largely underexplored. Previous attempts to utilize rational approximations have faced challenges, including high computational costs or reliance on polynomial intermediaries, limiting their full potential. This paper introduces TS-RatGNN, a novel spectral GNN featuring an explicitly-optimized rational filter. TS-RatGNN employs a distinctive two-stage framework: it sequentially applies numerator and denominator filters to the input signal. This streamlines the model architecture, facilitates efficient implementation, …

Leveraging Spectral Graph Properties For Unsupervised Cross-Domain Graph Classification

Defensive Publications Series

Unsupervised graph domain adaptation (UGDA) aims to transfer knowledge learned from labeled source graph datasets to unlabeled target graph datasets originating from a different distribution. Existing approaches often rely on spatial message-passing mechanisms, potentially overlooking the valuable information encoded in the spectral domain specific to UGDA tasks. This work begins with an empirical investigation revealing that low-frequency components of the graph spectrum tend to capture shared features across domains, whereas high-frequency components often represent domain-specific characteristics. Addressing the challenge of effectively utilizing these spectral insights, we introduce the Spectral Synergy Network (SSN), a novel framework for UGDA. SSN explicitly disentangles …

Navigating Extended Reality Maps Using Large Language Models, Tony (Tuấn) Nguyễn, Chunpo Wang, Hiro Tsujino, Stiven Morvan

Defensive Publications Series

Map applications in extended reality (XR map-apps) immersively display map data and features within extended reality (XR) environments. Currently, users navigate within an XR map-app using hand tracking and/or gaze tracking, which can be inconvenient. This disclosure describes techniques that leverage large language models (LLMs) to enable users to query an XR map-app using voice commands. With user permission, the user’s XR environment, including the user’s real environment (under XR passthrough) and the user’s virtual environment (e.g., immersive screen, audio, etc.) is shared with the LLM. A question from the user can be answered by the LLM in the context …

Product Documentation Updates Using Artificial Intelligence, Kayce Basques, Meggin Kearney

Defensive Publications Series

Keeping product documentation in synchronization with a product as the product is updated requires significant manual effort and can be prone to human error. This disclosure describes techniques to automatically update documentation when changes are made to a product or system covered by the documentation. Per the techniques, embeddings are generated for individual documents (or sections thereof) in the documentation corpus. As the product or system is updated, the corresponding changes are recorded, e.g., in an updated codebase for a software product, design and/or hardware specifications for a hardware product, etc. New embeddings are generated for the updated information based …

Analytic Hierarchy Process Based Interpretable Decision-Making Structure Of Gpt In Unified Developer Efficiency Modeling, Fuming Guo, Biju Abraham

Defensive Publications Series

The present disclosure relates to a method and a system for Analytic Hierarchy Process (AHP) based interpretable decision-making structure of Generative Pre-trained Transformer (GPT) in unified developer efficiency modeling. The method includes receiving and preprocessing a plurality of metrics from one or more data sources. The method further generates pairwise comparison scores for each metric of the plurality of metrics. Additionally, an AHP comparison matrix is constructed based on the generated comparison scores. The method further includes performing a consistency check on the comparison matrix to confirm the reliability of the pairwise comparisons. When inconsistencies are identified, the method includes …

Unizyme A Unified Protein Cleavage Site Predictor Enhanced With Enzyme Active-Site Knowledge

Defensive Publications Series

Enzyme-catalyzed protein cleavage is essential for many biological functions. Accurate predic tion of cleavage sites can facilitate various appli cations such as drug development, enzyme de sign, and a deeper understanding of biologica mechanisms. However, most existing models are restricted to an individual enzyme, which ne glects shared knowledge of enzymes and fails generalize to novel enzymes. Thus, we introduce a unified protein cleavage site predictor named UniZyme, which can generalize across diverse enzymes. To enhance the enzyme encoding for the protein cleavage site prediction, UniZyme em ploys a novel biochemically-informed model ar chitecture along with active-site knowledge of pro …

Probabilistic Graph Circuits Deep Generative Models For Tractable Probabilistic Inference Over Graphs

Defensive Publications Series

Deep generative models (DGMs) have recently demonstrated remarkable success in capturing complex probability distributions over graphs. Al though their excellent performance is attributed to powerful and scalable deep neural networks, it is, at the same time, exactly the presence of these highly non-linear transformations that makes DGMs intractable. Indeed, despite representing probability distributions, intractable DGMs deny probabilistic foundations by their inability to answer even the most basic inference queries with out approximations or design choices specific to a very narrow range of queries. To address this limitation, we propose probabilistic graph circuits (PGCs), a framework of tractable DGMs that provide …

Offline Model-Based Optimization Comprehensive Review

Defensive Publications Series

Offline optimization is a fundamental challenge in science and engineering, where the goal is to optimize black-box functions using only offline datasets. This setting is particularly relevant when querying the objective function is prohibitively expensive or infeasible, with applications spanning protein engineering, material discovery, neural architecture search, and beyond. The main difficulty lies in accurately estimating the objective landscape beyond the available data, where extrapolations are fraught with significant epistemic uncertainty. This uncertainty can lead to objective hacking (reward hacking)—exploiting model inaccuracies in unseen regions—or other spurious optimizations that yield misleadingly high performance estimates outside the training distribution. Recent advances …

Molground: A Benchmark For Molecular Grounding

Defensive Publications Series

Current molecular understanding approaches predominantly focus on the descriptive aspect of human perception, providing broad, topic level insights. However, the referential aspect—linking molecular concepts to specific structural components—remains largely unexplored. To address this gap, we propose a molecular grounding benchmark designed to evaluate a model’s referential abilities. We align molecular grounding with established conventions in NLP, cheminformatics, and molecular science, showcasing the potential of NLP techniques to advance molecular under standing within the AI for Science movement. Furthermore, we constructed the largest molecular understanding benchmark to date, compris ing 79k QA pairs, and developed a multi-agent grounding prototype as proof …

Cur-Gad: Calibrated Uncertainty And Risk Control For Graph Anomaly Detection

Defensive Publications Series

Graph Anomaly Detection (GAD) is critical in security-sensitive domains, yet faces reliability challenges: mis-calibrated confidence estimation (underconfidence in normal nodes, overconfidence in anomalies), adversarial vulnerability of derived confidence score under structural perturbations, and limited efficacy of conventional calibration methods for sparse anomaly patterns. Thus we propose CUR-GAD, a framework integrating statistical risk control into GAD via two innovations: (1) A Dual- Threshold Conformal Risk Control mechanism that provides theoretically guaranteed bounds for both False Negative Rate (FNR) and False Positive Rate (FPR) through providing prediction sets; (2) A Subgraph-aware Spectral Calibrator (SASC) that optimizes node representations through adaptive spectral filtering …

Hybrid Spectral-Spatial Learning For Robust Graph Anomaly Detection

Defensive Publications Series

Anomaly detection on graph data has garnered significant interest from both the academia and industry. In recent years, fueled by the rapid development of Graph Neural Networks (GNNs), various GNNs-based anomaly detection methods have been proposed and achieved good results. However, GNNs-based methods often assume that connected nodes have similar classes and features (homophily), leading to issues of class inconsistency and semantic inconsistency in graph anomaly detection, particularly when this assumption is violated. Existing methods have yet to adequately address these issues simultaneously, thereby limiting the detection performance of the model. Therefore, an anomaly detection method that consists of one …

Refining Multi-Relational Graph Anomaly Detection With Learned Structures And Scalable Meta-Path Aggregation

Defensive Publications Series

Graph Neural Networks (GNNs) have recently achieved remarkable success in various learning tasks involving graph-structured data. However, their application to multi-relational graph anomaly detection problems on real-world datasets presents several challenges that significantly hinder performance: the structural noise and inconsistencies inherent in real-world graph data, difficulties in aggregating information across multiple relation types, and the imbalanced class labels. To address these limitations, we introduce a graph structure learning layer designed to refine the original, noisy graph structure, enhancing the representation of node relationships. This enables our model to effectively handle inconsistencies and structural noise present in the original graph data …