Jared Dunnmon

Investigation of Lean Combustion Stability, Pressure Drop, and Material Durability in Porous Media Burners

The operational stability and thermal durability of combustion in two-zone porous media burners (PMBs) is examined experimentally and computationally. Long-term material durability tests at constant and cycled on-off conditions are performed, along with a characterization of combustion stability, pressure drop and pollutant emissions for a range of equivalence ratios, mass flow rates, and burner setups. Experimental thermocouple temperature measurements and pressure drop data are presented and compared to results obtained from one-dimensional volume-averaged simulations. Experimental and model results show good agreement for temperature profiles and pressure drop evaluated using the Darcy-Forchheimer equation with Ergun’s relations. Enhanced flame stability is observed for burners with Yttria-stabilized Zirconia Alumina (YZA) upstream and Silicon Carbide (SiC) in the downstream combustion zone. Measurements of product gas concentrations illustrate highest emissions of CO at conditions close to flash-back and, as expected, higher NOx emissions with increasing equivalence ratios. ∗Address all correspondence to this author. NOMENCLATURE Di j Species i binary diffusion coefficient (m/s) MFR Mass flux rate (kg/m2s) Pe Peclet number (Pe = SLdp,e f f ρgcg λg ) SL Laminar flame speed (m/s) Xi Species i mole fraction Yi Species i mass fraction c Specific heat capacity (J/KgK) dp Pore diameter (m) hv Volumetric heat transfer coefficient (W/m3K) ṁ Mass flow rate (kg/s) q̇ Heat release rate (W/m3) u Volume-averaged fluid velocity (m/s) ε Porosity λ Thermal conductivity (W/mK) κ Radiative heat extinction coefficient (W/m2K) Ω Scattering albedo ω̇i Species i production rate per unit volume (kg/m3s) φ Equivalence ratio ρ Density (kg/m3) σ Stefan-Boltzmann constant (W/m2K4) 1 Copyright

Snorkel MeTaL: Weak Supervision for Multi-Task Learning

Many real-world machine learning problems are challenging to tackle for two reasons: (i) they involve multiple sub-tasks at different levels of granularity; and (ii) they require large volumes of labeled training data. We propose Snorkel MeTaL, an end-to-end system for multi-task learning that leverages weak supervision provided at multiple levels of granularity by domain expert users. In MeTaL, a user specifies a problem consisting of multiple, hierarchically-related sub-tasks---for example, classifying a document at multiple levels of granularity---and then provides labeling functions for each sub-task as weak supervision. MeTaL learns a re-weighted model of these labeling functions, and uses the combined signal to train a hierarchical multi-task network which is automatically compiled from the structure of the sub-tasks. Using MeTaL on a radiology report triage task and a fine-grained news classification task, we achieve average gains of 11.2 accuracy points over a baseline supervised approach and 9.5 accuracy points over the predictions of the user-provided labeling functions.

Weakly supervised classification of rare aortic valve malformations using unlabeled cardiac MRI sequences

Recent releases of population-scale biomedical repositories such as the UK Biobank have enabled unprecedented access to prospectively collected medical imaging data. Applying machine learning methods to analyze these data holds great promise in facilitating new insights into the genetic and epidemiological associations between anatomical structures and human health. However, the majority of these imaging data are unlabeled and deriving insights is hindered by the cost of manually annotating data at sufficient scale to train state-of-the-art deep learning models. In this work, we develop a weakly supervised deep learning model for Bicuspid Aortic Valve (BAV) classification using up to 4,000 unlabeled cardiac MRI sequences, comprising a total of 120,000 images. Instead of requiring manually labeled training data, weak supervision relies on noisy heuristic functions defined by domain experts to automatically generate large-scale, imperfect training sets. By leveraging new theoretical work on coping with label noise, models can use weaker supervision sources than was previously possible. In our BAV models, this approach substantially outperforms a traditional supervised baseline trained on hand-labeled data alone, with a 64% improvement in mean F1 score (37.8 to 61.4) on held out test data. In a validation experiment using 9,230 individuals with MRIs and long-term outcome data from the UK Biobank, applying the best-performing BAV classification model identified a subset of individuals with a 1.8-fold increase in risk of a major adverse cardiac event (p <0.001). This work formalizes the first deep learning baseline for aortic valve classification and outlines a general strategy for using weak supervision to analyze large collections of unlabeled medical images. Author summary We developed a deep learning model for Bicuspid Aortic Valve (BAV) classification using up to 4,000 unlabeled cardiac MRI sequences, comprising a total of 120,000 images. Instead of requiring manually labeled training data, as is typical in machine learning, our approach relies on noisy heuristic functions defined by domain experts to automatically generate large-scale, imperfect training sets. In our experiments, this approach substantially outperforms a traditional supervised baseline trained on hand-labeled data alone. In a validation experiment using 9,230 individuals with MRIs and long-term outcome data from the UK Biobank, applying the best-performing BAV classification model identified a subset of individuals with a 1.8-fold increase in risk of a major adverse cardiac event. This work formalizes the first deep learning baseline for aortic valve classification and outlines a general strategy for using weak supervision to analyze large collections of unlabeled medical images.