More Like this

1. Compressive sensing with un-trained neural networks: Gradient descent finds the smoothest approximation

   Heckel, Reinhard ; Soltanolkotabi, Mahdi ( July 2020 , International Conference on Machine Learning)

   Un-trained convolutional neural networks have emerged as highly successful tools for image recovery and restoration. They are capable of solving standard inverse problems such as denoising and compressive sensing with excellent results by simply fitting a neural network model to measurements from a single image or signal without the need for any additional training data. For some applications, this critically requires additional regularization in the form of early stopping the optimization. For signal recovery from a few measurements, however, un-trained convolutional networks have an intriguing self-regularizing property: Even though the network can perfectly fit any image, the network recovers a natural image from few measurements when trained with gradient descent until convergence. In this paper, we provide numerical evidence for this property and study it theoretically. We show that---without any further regularization---an un-trained convolutional neural network can approximately reconstruct signals and images that are sufficiently structured, from a near minimal number of random measurements. 

   more » « less
2. Compressive sensing with un-trained neural networks: Gradient descent finds the smoothest approximation

   Heckel, Reinhard ; Soltanolkotabi, Mahdi ( January 2020 , International Conference on Machine Learning)

   Un-trained convolutional neural networks have emerged as highly successful tools for image recovery and restoration. They are capable of solving standard inverse problems such as denois- ing and compressive sensing with excellent results by simply fitting a neural network model to measurements from a single image or signal without the need for any additional training data. For some applications, this critically requires additional regularization in the form of early stopping the optimization. For signal recovery from a few measurements, however, un-trained convolutional networks have an intriguing self-regularizing property: Even though the network can perfectly fit any image, the network recovers a natural image from few measurements when trained with gradient descent until convergence. In this paper, we provide numerical evidence for this property and study it theoretically. We show that—without any further regularization—an un-trained convolutional neural network can approximately reconstruct signals and images that are sufficiently structured, from a near minimal number of random measurements 

   more » « less
3. K. Le, T. Butterfield, K.E. Kelly, Citizen scientist engagement in air quality monitoring, Proc. ASEE Annual Meeting, June 23 – 27, Salt Lake City, UT.

   Le, K. ; Butterfield, T. ; Kelly, K.E. ( July 2018 , Review & directory - American Society for Engineering Education)

   Citizen scientist efforts, wherein members of the public who are not professional scientists participate in active research, have been shown to effectively engage the public in STEM fields and result in valuable data, essential to answering pressing research questions. However, most citizen scientist efforts have been centered in colleges of science, and a limited number have crossed into research areas important to chemical engineering fields. In this work we report on the results of a project to recruit high school and middle school students across Utah’s Salt Lake Valley as citizen scientists and potential engineering students who work in partnership with chemical engineering researchers in an effort to create a distributed online network of air quality sensors. Middle and high school students were trained by undergraduate mentors to monitor and maintain their own outdoor air quality sensor with the help of teaching materials that were co-developed with Breathe Utah, a local community group concerned with air quality. With the help of these tailored teaching modules, students learned about the science behind air quality research and the difficulties common to physical measurements to better prepare them to analyze their data. Once trained, students are expected to become semi-independent researchers in charge of monitoring and maintaining their piece of a larger air quality map. We describe in this work the hurdles inherent in citizen science engagement within a chemical engineering research program and the means to address them. We describe successful means of engaging classrooms, training citizen scientists, obtaining faculty buy-in within the confines of state curricular demands, and addressing school administration concerns. With this model, we have directly engaged over 1,000 high school and over 3,000 middle school students. The project has resulted in a growing network of citizen-maintained sensors that contributes to a real-time air quality map. Student scientists may also use the sensors to participate in active research or conduct science fair projects. Student response to this citizen scientist project, where it may be measured, has been enthusiastic and almost wholly positive. 

   more » « less
4. Comparison of Spectroscopic Techniques for Determining the Peroxide Value of 19 Classes of Naturally Aged, Plant-Based Edible Oils

   https://doi.org/10.1177/0003702821994500  

   Ottaway, Joshua_M ; Chance_Carter, J. ; Adams, Kristl_L ; Camancho, Joseph ; Lavine, Barry_K ; Booksh, Karl_S ( April 2021 , Applied Spectroscopy)

   The peroxide value of edible oils is a measure of the degree of oxidation, which directly relates to the freshness of the oil sample. Several studies previously reported in the literature have paired various spectroscopic techniques with multivariate analyses to rapidly determine peroxide values using field portable and process instrumentation; those efforts presented “best-case scenarios” with oils from narrowly defined training and test sets. The purpose of this paper is to evaluate the use of near- and mid-infrared absorption and Raman scattering spectroscopies on oil samples from different oil classes, including seasonal and vendor variations, to determine which measurement technique or combination thereof is best for predicting peroxide values. Following peroxide value assays of each oil class using an established titration-based method, global and global-subset calibration models were constructed from spectroscopic data collected on the 19 oil classes used in this study. Spectra from each optical technique were used to create partial least squares regression calibration models to predict the peroxide value of unknown oil samples. A global peroxide value model based on near-infrared (8 mm optical path length) oil measurements produced the lowest RMSEP (4.9), followed by 24 mm optical path length near infrared (5.1), Raman (6.9) and 50 µm optical path length mid-infrared (7.3). However, it was determined that the Raman RMSEP resulted from chance correlations. Global peroxide value models based on low-level fusion of the NIR (8 and 24 mm optical path length) data and all infrared data produced the same RMSEP of 5.1. Global subset models, based on any of the spectroscopies and olive oil training sets from any class (pure, extra light, extra virgin), all failed to extrapolate to the non-olive oils. However, the near-infrared global subset model built on extra virgin olive oil could extrapolate to test samples from other olive oil classes. This work demonstrates the difficulty of developing a truly global method for determining peroxide value of oils.

    

   more » « less
5. Implementing the Elements of Course-Based Undergraduate Research Experiences (CUREs) in a First-Year Undergraduate Chemistry Laboratory with Bioremediation Relevance

   https://doi.org/10.1021/acs.jchemed.2c00360  

   Silsby, Carson ; McCormack, Roslyn ; Roll, Mark F. ; Moberly, James G. ; Waynant, Kristopher V. ( July 2022 , Journal of Chemical Education)

   General chemistry laboratories are a core requirement for nearly all STEM (Science, Technology, Engineering, and Mathematics) majors and have the greatest breadth in disciplines and audience of any STEM course at a university. A bioremediation Course-based Undergraduate Research Experience (CURE) for first-year undergraduate students was developed to capture and engage student interest for this diverse group. In this multiweek laboratory exercise, students joined an NSF-funded research project designed to enhance the bioremediation of chlorinated aliphatic hydrocarbons. Students explored various biocompatible polymer blends and cross-linkers for encapsulation to create protection for bioremediation microbes. In this “guided research” model, students constructed the measurement apparatus, made hydrogel blends, and then monitored the diffusion of acid via pH measurements using a custom instrument. Herein, we describe how CURE elements were implemented within the bioremediation research experience culminating in student teams presenting posters at our university’s undergraduate research symposium. An open-laboratory format facilitated an active research group experience and recitation “group meeting” provided flexibility and needed time for reflection and discussion. Student survey data and course evaluations indicated that students saw value in this genuine research experience and enjoyed the freedom and time to practice and hone skills as both a scientist and teammate in a laboratory setting. 

   more » « less