Digital Commons Network^™

Challenges And Best Practices Of Regional Innovation Ecosystems, Mahnaz Asgari Sooran

Miners Solving for Tomorrow Research Conference

Many places around the world are developing regional innovation ecosystems to spur regional economic. Many innovation ecosystems fail or barely maintain their initial momentum after a few years. In this paper, we identified challenges and best practices facing innovation ecosystems interviews with the innovation ecosystem stakeholders. We employed the MIT REAP (Regional Entrepreneurship Acceleration Program) model to categorize the five main stakeholders of the innovation ecosystem: entrepreneurs, universities, industry, risk capital, and government. Our initial results indicated the following: (a) Access to capital and talented workforce; (b) Stakeholder discovery process is one of the critical areas to understand the basic …

Introducing Gridtrees For Streaming, Scalable Hierarchical Data Visualization, Nathan Tibbetts

Miners Solving for Tomorrow Research Conference

File browsing in the consumer sphere has seen very little advancement in recent years, although attempts have been made to improve upon it. With the goal of multi-level visual file browsing in mind, we present a prototype GridTree, a dynamic, recursive, stable, spatial layout data structure based on subdividing grids, represented as a hierarchy of maps whose granularity increases with depth. We motivate this work with characteristics we have identified as important for viability of a multi-level file-browser, which have become our design goals. We touch on the underlying logic of a GridTree and identify its complexity. We briefly discuss …

Simulating Thermal Diffusion Through Image-Derived Microstructures Of Ceramic Matrix Composites, Matik Heskin

Miners Solving for Tomorrow Research Conference

Thermal energy transport in materials can be effectively modeled using finite element software such as COMSOL. Experimentally measured or NIST–JANAF thermal conductivity data can be used to represent material behavior within these simulations. For heterogeneous or composite materials, effective properties are often approximated using rule-of-mixtures calculations. However, this overlooks the nuanced effects caused by complex microstructural geometry. To address this limitation, imaging and coding tools such as MATLAB can be used to process scanning electron microscopy (SEM) images. By thresholding the images to distinguish constituent materials, a representative mesh can be generated and imported into an FEM program. This approach …

Thermal Mirage: Towards Robust Thermal Perception Via Gan-Guided Diffusion, Nuzaer Omar

Miners Solving for Tomorrow Research Conference

Thermal object detection systems are critical for safety sensitive applications due to their reliability under adverse conditions. However, existing robustness evaluations in thermal domain primarily focuses on physical or sensor-level perturbations, overlooking vulnerabilities from semantically realistic scene and object manipulations. We introduce Thermal Mirage, a generative framework that leverages GAN-guided diffusion to expose weaknesses in thermal detectors through controlled object and context level perturbations. Our approach learns class-conditional thermal priors via a GAN and uses diffusion to transform object appearances into ambiguous or low-saliency patterns. Simultaneously, a context module degrades scene conditions by simulating harsher night environments. Integrated with a …

Human Intention Prediction Using Cnn And Lstm Networks In Physical Human–Robot Interactions, Khosro Ghorbani Zadeh

Miners Solving for Tomorrow Research Conference

Advances in robotics and artificial intelligence have increased expectations for interactive robots in eldercare and assistive applications. A key challenge in creating safe and effective systems is accurately recognizing human intent and translating it into meaningful commands for robots to follow. Traditional physics-based models often fail to fully represent human force interactions due to their dynamic nature, while neural networks provide a promising alternative for predicting force-movement intentions. Multi-layer perceptrons (MLPs) show potential, however, they struggle with temporal dependencies and generalization. Long Short-Term Memory (LSTM) networks and Convolutional Neural Networks (CNNs) help address these limitations. This study compares these architectures …

Ai-Driven Frameworks For Mitigating Rockfall Hazards, Unstable Ground Condition And Improving Safety In Underground Mine Through Thermal Imaging, Akhrorbek Narmatov

Miners Solving for Tomorrow Research Conference

Scaling, the removal of loose rocks from excavation walls to prevent rockfalls is a critical safety procedure in underground mining operations. Approximately a quarter of fatal mining accidents are due to rockfalls. Detecting unstable rocks during scaling and assessing ground stability remain challenging due to limited visibility, dust, and environmental conditions.

Excavation disturbs the in-situ stress state, causing stress redistribution, fracture propagation, and the formation of discontinuities that lead to loose rock blocks. Ventilation airflow affects heat transfer at exposed surfaces, amplifying temperature contrasts between intact rock and fractured blocks associated with instability.

Current practices rely on manual inspection, requiring …

Robust Federated Learning With Strategic Adversaries, Manoj Twarakavi

Miners Solving for Tomorrow Research Conference

Federated Learning (FL) leverages the intelligence of untrusted distributed devices through collaborative training. This makes the training process susceptible to malicious behavior. Existing defense mechanisms largely consider an adversary who attacks a proactive FL server without any adaptability. However, they overlook the presence of a strategic adversary. To address this challenge, our work proposes a Robust Game-theoretic framework where the adversary is both strategic and is equipped with the capability of performing large-scale poisoning attacks.

Harnessing Wavefront Shaping Control For Sensing Applications, Pablo Jara

Miners Solving for Tomorrow Research Conference

Diffuse optical tomography (DOT) and functional near-infrared spectroscopy (fNIRS) enable deep, non-invasive sensing in biological tissue but are fundamentally limited by the photon budget - most injected light is lost to scattering before reaching the detector. Wavefront shaping (WFS) can enhance signal strength inside scattering media via interference, but the conventional diffusion-based sensitivity model breaks down under coherent illumination. We develop a microscopic theory of optical sensitivity that captures interference effects neglected by diffusion theory. We prove analytically that the microscopic and diffusive descriptions coincide under random illumination and identify WFS strategies that enhance sensitivity beyond this limit. The maximum …

Variational Data Assimilation With Steepest Descent Method For Coupled Time-Dependent Stokes-Darcy Model With Bjsj Interface Condition, Yafang Hei

Miners Solving for Tomorrow Research Conference

Variational data assimilation (VDA) determines the initial condition of a dynamical system by minimizing the mismatch between model predictions and observed data. This work studies VDA for the time-dependent Stokes–Darcy system with the BJSJ interface condition. The problem is formulated as a PDE-constrained optimization problem, and the first-order optimality system is derived using the Gâteaux derivative and adjoint variables. A steepest descent method is applied for efficient computation. Spatial and temporal discretizations are carried out using the finite element method and backward Euler scheme, respectively. Numerical results confirm accuracy and convergence.

Dispersive Shock And Rogue Waves In Two-Dimensional Quantum Droplets, Farhana Bristy

Miners Solving for Tomorrow Research Conference

Quantum droplets, are liquid type configurations stabilized by the balance between attractive mean-field interactions and repulsive quantum fluctuations. They provide highly flexible platforms for the quantum simulation of hydrodynamic phenomena using ultracold gases. Here, we explore the nonlinear quantum dynamics of two-dimensional quantum droplets under Riemann initial conditions. The steepness of the latter facilitates the emergence of radially symmetric dispersive shock waves (DSWs) when quantum fluctuations dominate. In particular, the ensuing DSWs travel from the potential edges toward the center where they collide and through their interference high amplitude spatially localized rogue wave structures emerge. In contrast, tuning the interactions …

Distributed Control Plane For Cross-Silo Federated Learning, Rabin Pandey

Miners Solving for Tomorrow Research Conference

Cross-silo federated learning (FL) trains shared models across geographically distributed institutions without centralizing raw data, but its reliance on wide-area networks makes round completion time highly sensitive to heterogeneous link conditions, congestion, and straggler clients. Existing SDN-based FL frameworks address this through centralized traffic engineering, an approach that breaks down when silos span independent administrative domains where no single entity can maintain complete, timely network knowledge. We replace the centralized control plane with a fully distributed overlay in which each silo gateway operates as an equal peer, continuously probing local links, exchanging EWMA-smoothed metrics via bounded gossip, and computing least-cost …

Transition Metal Chalcogenide As Multifunctional Biosenser To Detect Neurochemicals, Amitav Sen

Miners Solving for Tomorrow Research Conference

The development of highly sensitive real-time sensing platforms for selectively detecting dopamine (DA) and norepinephrine (NE) is of significant clinical importance since these are closely associated with several neurodegenerative disorders. Herein, we report copper telluride (Cu₂Te) nanostructure as a bifunctional electrocatalyst for the simultaneous detection of dopamine and norepinephrine. Cu₂Te exhibits excellent electrocatalytic activity toward both DA and NE oxidation, as demonstrated by detailed electrochemical measurements. High sensitivities of 24 and 40 µA cm⁻² µM⁻¹ for DA and NE, respectively, were obtained with a wide linear detection range of 10–200 nM and low limit of detection (LOD), 31 nM for …

Toward Sustainable 3d Printable Cementitious Materials: Low-Carbon Design With Fiber Reinforcement, Nima Mahmoudzadeh Vaziri

Miners Solving for Tomorrow Research Conference

3D printing of cementitious materials, as an emerging construction technology, faces challenges in both mechanical performance and sustainability. Conventional mixture designs rely on high cement contents, leading to increased CO2 emissions, while printed elements often exhibit brittle behavior with limited crack resistance and reduced post-cracking performance. This study investigates the development of low-carbon 3D printed cementitious composites using environmentally friendly supplementary cementitious materials (SCMs) and fillers, combined with fiber reinforcement to enhance mechanical performance. Locally available SCMs and fillers were incorporated to significantly reduce cement content (>40%) and improve sustainability. Fiber reinforcement was introduced to mitigate brittleness and enhance …

Scenario-Based Pareto Analysis For Multi-Objective Optimization Of Process Parameters To Improve Life Cycle Impacts: Case Study Of Ge Production, Dennis Dadzie

Miners Solving for Tomorrow Research Conference

Growing demand for critical minerals increases the need for clear evidence on the environmental burdens of their production. Life‑cycle assessment (LCA) quantifies impacts, but it does not by itself show how to navigate competing environmental performance improvement goals. We examine 63 process configurations for single‑crystal germanium made via chlorinated distillation using five indicators: global warming potential (GWP), water consumption (WC), fine particulate matter (FPM), terrestrial acidification (TA), and cumulative energy demand (CED). Across scenarios, WC is negatively correlated with GWP, weakly related to FPM, TA, and CED, while the other four indicators are strongly positively correlated. Pareto analysis highlights a …

Assessing Radon And Radon Progeny Deposition On Simulated Lung Filter System: Effects Of Cigarette And E-Cigarette Use Based On The Icrp Model, Manuela Isabel Arenas Alvarez

Miners Solving for Tomorrow Research Conference

This project investigates the potential interaction between radon exposure and e-cigarette aerosols, focusing on how radon progeny deposit in the lungs under combined conditions. An experimental system was developed to simulate regional lung deposition based on the International Commission on Radiological Protection (ICRP) model. The setup uses a filter-based design and a custom modular atomizer to generate stable aerosols, with four commercial plastic filters evaluated. Initial results demonstrate reproducible filtration behavior and a total deposition pattern matching ICRP predictions within 21% error. Ongoing work aims to improve agreement with regional deposition curves and enable combined radon–aerosol exposure experiments. This platform …

Measuring The Influence Of Ai Decision Support On Organ Procurement Coordinators’ Workflow, Amaneh Babaee

Miners Solving for Tomorrow Research Conference

Approximately 30% of deceased-donor kidneys remain unused, representing missed opportunities for transplantation. One potential solution is identifying hard-to-place kidneys earlier in the allocation process to reduce late-stage reassessment and allocation inefficiencies. Over a four-year project, our team developed an artificial intelligence (AI) algorithm that provides Organ Procurement Organization (OPO) coordinators with an AI score indicating the likelihood that a kidney will be transplanted. This study evaluates the impact of this AI score on kidney placement decisions. Using a pre–post experimental design, participants will first evaluate four donor kidneys without AI support. After watching a training video explaining the AI score, …

Preventive Maintenance Scheduling Using Artificial Intelligence And Decision Support Agent, Samiksha Aryal

Miners Solving for Tomorrow Research Conference

Preventive Maintenance models have traditionally relied on a time-based maintenance model, which uses fixed statistical distribution to represent the time-to-failure (TTF). The requirement of data-driven models and automated decision-making systems have become essential for modern manufacturing systems. The use of fixed statistical distribution limits the ability of existing models in producing solutions in real-time. Our model overcomes these limitations by implementing an empirical distribution to represent TTF. The empirical distribution is generated using a neural network which eliminates noise from the raw maintenance log. Renewal Reward Theorem (RRT) is implemented to efficiently provide maintenance threshold in real time. The use …

Robustness Of Fuzzy Artmap To Adversarial Attacks And Progressive Adversarial Training For Streaming Learning, Shane Cairns

Miners Solving for Tomorrow Research Conference

Incremental learners deployed on streaming data must remain robust to evolving adversarial perturbations, yet most adversarial-robustness studies assume offline multi-epoch training with repeated access to historical data. We investigate adversarial robustness in Fuzzy ARTMAP, a prototype-based Adaptive Resonance Theory model that supports single-pass learning without replay. We propose WB-Softmax, a differentiable relaxation that aggregates category-level activations into class-level scores for gradient-based attacks. WB-Softmax PGD achieves 89–100% attack success on vanilla models, exceeding transfer and query-based baselines. We then study adversarial training under true streaming constraints by comparing offline versus online adversarial example generation and standard versus selective updates. Offline adversarial …

2nd Annual Miners Solving For Tomorrow Research Conference - Graduate Schedule, Antai Dong, Connor Bell, Dennis Dadzie, Joshua Gary, Adebayo Olayinka Oke, Deeshani Mitra, Yejun Kim, Josiah Mcdermott, Samiksha Aryal, Nicholas Benner, Timothy Ennis, Maryam Sharifi Paroushi, Sujan Maharjan, Md Saiduzzaman, Hari Dhital, Manuela Isabel Arenas Alvarez, Eyuel A Getahun, Lucas Greiner, Amaneh Babaee, Shane Cairns, Nazish Khalid, Joseph Walton, Jie Shi, Ehsan Asheghianamiri, Mariam Elazhary, Makuach James Makeny Panther Athach, Rasman Mubtasim Swargo, Shiva Kumar Goud Kasani, Pablo Jara, Sara Mccauley, Hartzell Weston, Al Mojahid Afridi, Matik Heskin, Nima Mahmoudzadeh, Effat Eskandari, Amitav Sen, Rabin Pandey, Amirhossein Habibi, Nuzaer Omar, Mizanur Rahman Jewel, Nathan Tibbetts, Manoj Twarakavi, Farhana Bristy, Akhrorbek Narmatov, Mahnaz Asgari Sooran, Yafang Hei, Nathan Roberts, Khosro Ghorbani Zadeh

Miners Solving for Tomorrow Research Conference

No abstract provided.

Assessing The Effects Of Scale-Dependent Representation Of River Networks On Hydrologic Prediction, Hari Dhital

Miners Solving for Tomorrow Research Conference

Accurate hydrologic prediction at relevant spatial scales is essential for flood risk management. This study explores how spatial scale affects model parameters and simulation outcomes in the Hillslope Link Model (HLM). This was motivated by HLM implementation within the Next Generation Water Resources Modeling (NextGen) framework for the U.S. National Water Model. Since NextGen alters HLM's spatial scale from hillslopes to catchments, we assess how this change affects runoff generation and routing. We implement HLM at both scales using identical, predetermined parameters without calibration to isolate spatial discretization effects. Using approximately ten years of precipitation and streamflow data from watersheds …

Evaluation Of National Water Model Long-Range Streamflow Forecasts In Missouri, Sujan Maharjan

Miners Solving for Tomorrow Research Conference

Long-range river stage forecasts are increasingly used to support barge scheduling along the Missouri and Mississippi River navigation corridor. The reliability of these forecasts over extended lead times remains uncertain, particularly in regulated systems influenced by upstream dam operations. This study evaluates the lead-time-dependent performance of National Water Model long-range forecasts at eleven U.S. Geological Survey stations from Rulo to Thebes during 2019–2024. Results show that forecast skill does not decline smoothly with time but exhibits station-specific dips at lead times corresponding to hydraulic travel from major control structures. These periods coincide withstage bias and dispersion, indicating elevated operational uncertainty. …

Evaluation Of Channel Routing Module In The Nextgen National Water Model, Md Saiduzzaman

Miners Solving for Tomorrow Research Conference

The Next Generation Water Resources Modeling (NextGen) framework has a modular structure that enables the integration of multiple hydrologic models with channel routing processes. Within NextGen, the T-Route model performs channel routing using vector-based river networks and supports a hybrid routing approach within a single watershed. This study evaluates the hydraulic diffusive wave routing method implemented in T-Route and compares its performance with the operational Muskingum-Cunge routing used in the National Water Model. The primary objective is to assess the operational capability of streamflow data assimilation across different routing methods using an extensive network of streamflow observations. The routing models …

Controlled Gelation Of Co₂-Resistant Polymer Systems For Improved Conformance In Co₂-Eor, Maryam Sharifi Paroushi

Miners Solving for Tomorrow Research Conference

This presentation introduces a CO₂-resistant hydrogel developed to control fluid movement and improve CO₂ injection performance in subsurface reservoirs for Enhanced Oil Recovery (EOR) and Carbon Capture and Storage (CCS). Conventional polymer gels often lose strength in CO₂-rich environments, resulting in weak plugging, early breakdown, and ineffective flow control. To address this limitation, a sulfonated HPAM-based gel system crosslinked with chromium was developed. The effects of polymer concentration, molecular weight, crosslinker ratio, ionic strength, and pH were investigated to better understand gelation behavior and performance. The formulation exhibited controlled gelation, strong resistance to temperature and CO₂, and minimal shrinkage over …

Empirical Evaluation Of Policy-Based Reinforcement Learning For Dynamic Service Control In An M/M/1 Queue, Joseph Walton

Miners Solving for Tomorrow Research Conference

While reinforcement learning has been increasingly applied to stochastic control, limited work examines policy-based methods in queuing environments modeled as semi-Markov decision processes (SMDP). This study investigates how policy-based reinforcement learning (RL) algorithms perform when applied to service rate control in an M/M/1 queue, a common queuing model for manufacturing and service systems. The problem is formulated as an SMDP in which decisions occur at each new service, allowing an agent to select different service rates from a finite set of speeds, aiming to minimize an objective function that manages system congestion and energy costs. Three policy-based reinforcement learning algorithms, …

Structural Analysis On Iron Phosphate Glasses After Dechlorination Of Mixed Chloride Waste Forms, Lucas Greiner

Miners Solving for Tomorrow Research Conference

Molten salt reactors (MSRs) require durable waste forms for immobilizing halide-rich salts. This study examines iron phosphate glasses derived from a simple salt mixture (SSM), focusing on the effect of Fe₂O₃ additions (2.5–10 wt%) on structure and phase stability. Intermediates were processed at 300–600 °C and vitrified at 1050 °C. Raman spectroscopy and HPLC show that increasing iron content depolymerizes the phosphate network, shifting from Q² toward Q¹ species and reducing chain length. XRD confirms amorphous behavior for ≤7.5 wt% Fe₂O₃, while 10 wt% shows crystallization, with KFeP₂O₇ observed in samples processed at 400 °C. Despite reduced connectivity, iron enhances …

Direct Energy Cascade Of 2d Quantum Turbulence In Supersolids, Lukas Farthing

Miners Solving for Tomorrow Research Conference

The non-equilibrium turbulent response of periodically driven three-dimensional ultracold dipolar gases is induced by an external time-dependent ring potential. To model this system and monitor its non-equilibrium quantum dynamics, we invoke an extended Gross-Pitaevskii framework containing the first-order quantum correction to the mean-field energy functional. The shape of the external perturbing potential is chosen to trigger angular roton excitations, driving the supersolid configuration out of equilibrium and attaining a turbulent state. Following the generation of shallow vortical defects in the bulk, a direct cascade front manifests, transporting energy from larger to smaller length scales. This leads to a non-equilibrium quasi-steady …

Triple Active Bridge Implementation And Control, Nehemiah Milton

Miners Solving for Tomorrow Research Conference

The Triple Active Bridge (TAB) is a three-port power converter that allows power flow in both directions with galvanic isolation. This has a wide range of applications, like high-frequency DC-DC conversion, electric vehicles, microgrids and renewable energy systems. To control the TAB during operation, two phase shift parameters between the bridges are adjusted. In this presentation, I will showcase the software and simulation optimizations which extend previous work by reproducing hardware results which align with simulations. I will also go over the knowledge I’ve gained and the skills acquired through this project.

Cytokinesis And Dbf2, Katharine Gray

Miners Solving for Tomorrow Research Conference

The Mitotic Exit Network (MEN) is a signaling pathway that allows a dividing cell to exit cytokinesis. Dbf2 is a MEN protein kinase in yeast, which provides a simplified model for studying this pathway. By altering Dbf2’s activity through phosphorylation mutants, its effect on cytokinesis can be studied. Transformed cells with a mutant and degron Dbf2 allele are prompted to destroy the degron Dbf2 protein, leaving only mutated Dbf2 behind. This allows for clear analysis of the effect of the phosphorylation mutant on division. Fluorescence microscopy can be used to visualise this degradation, and the impact on actin ring formation …

Vortex Generation In Dipolar Supersolids, Jacob Harl

Miners Solving for Tomorrow Research Conference

Ultracold atomic gases offer highly tunable platforms for exploring complex quantum many-body phenomena. In addition, dipolar quantum gases of magnetic atoms featuring long-range anisotropic interactions are exquisite systems for realizing exotic phases-of-matter in an experimentally controlled way. A prototypical example is the supersolid state which simultaneously exhibits frictionless flow of superfuids and crystalline density modulation of solids. In this realm, we investigate the dynamical generation of vortex topological defects by imprinting a suitable phase jump in quasi-two-dimensional dipolar supersolids. This protocol enables the nucleation of a dark soliton which consecutively becomes unstable via the eponymous snake instability due to its …

Elucidating Complex H Behavior In Amorphous Oxide Semiconductors By Machine Learning, Lucas Ethington

Miners Solving for Tomorrow Research Conference

In this project, the disordered oxide structures will be mapped by a machine-learning algorithm (e.g., HDBSCAN) to identify characteristic behaviors of the proton across different material densities and defect types. This should help identify the under-coordinated, highly distorted, weakly-bonded, and dynamically unstable atoms to predict the most probable H locations. This fast and accurate prediction of energetically favorable H distribution will enable a reliable and fast screening of a large number of AOSs with variable cation and/or anion compositions. The approach will help find AOSs with suppressed numbers of M-OH defects (that form deep electron traps, limiting the number of …