Search for: All records

Disposal of industrial wastewater and activities such as CO2 depend on pressure conditions within deep geologic reservoirs. Injection and storage are also associated with induced seismicity, suggested to result from reservoir compartmentalization and leakage into faults. To understand subsurface pressure conditions within a major regional disposal reservoir, the carbonate Arbuckle Group of Oklahoma, we monitored the water levels in 15 inactive injection wells. The wells were monitored at 30-second intervals, with eight wells monitored since September 2016, and an additional seven from July 2017. All of the wells were monitored until early March 2020. Since 2016, well levels decreased in 3 of the 15 wells (a.k.a. hydraulic head), proportional to near-borehole fluid pressure even considering decreasing regional injection volumes during this period. The well pressures respond to three types of perturbations: (i) gravitational fkuctuations (a.k.a. solid-earth tides) (ii) distal and proximal earthquakes, and (iii) injections into nearby wells. Parameterization of tidal responses illustrates that the near wellbore environments have negative fluid flux (i.e. are leaking). Earthquakes cause differing pressure responses from well to well, with some highly sensitive to proximal events, some to distal events, and some apparently insensitive. Injections have variable impacts in some cases masking tidal and earthquake pressure signals. Collectively, there appears to be a threshold injection rate above which well pressure becomes less sensitive to the volume of injections within 15 km. Multi-scale geological structure and temporal permeability changes are likely controlling the pressure field, indicating leakage of fluids across the system. 

Convolutional neural networks (CNNs), a class of deep learning models, have experienced recent success in modeling sensory cortices and retinal circuits through optimizing performance on machine learning tasks, otherwise known as task optimization. Previous research has shown task-optimized CNNs to be capable of providing explanations as to why the retina efficiently encodes natural stimuli and how certain retinal cell types are involved in efficient encoding. In our work, we sought to use task-optimized CNNs as a means of explaining computational mechanisms responsible for motion-selective retinal circuits. We designed a biologically constrained CNN and optimized its performance on a motion-classification task. We drew inspiration from psychophysics, deep learning, and systems neuroscience literature to develop a toolbox of methods to reverse engineer the computational mechanisms learned in our model. Through reverse engineering our model, we proposed a computational mechanism in which direction-selective ganglion cells and starburst amacrine cells, both experimentally observed retinal cell types, emerge in our model to discriminate among moving stimuli. This emergence suggests that direction-selective circuits in the retina are ecologically designed to robustly discriminate among moving stimuli. Our results and methods also provide a framework for how to build more interpretable deep learning models and how to understand them. 

Convolutional neural networks (CNN) are an emerging technique in modeling neural circuits and have been shown to converge to biologically plausible functionality in cortical circuits via task-optimization. This functionality has not been observed in CNN models of retinal circuits via task-optimization. We sought to observe this convergence in retinal circuits by designing a biologically inspired CNN model of a motion-detection retinal circuit and optimizing it to solve a motion-classification task. The learned weights and parameters indicated that the CNN converged to direction-sensitive ganglion and amacrine cells, cell types that have been observed in biology, and provided evidence that task-optimization is a fair method of building retinal models. The analysis used to understand the functionality of our CNN also indicates that biologically constrained deep learning models are easier to reason about their underlying mechanisms than traditional deep learning models. 

Convolutional neural networks (CNN) are an emerging technique in modeling neural circuits and have been shown to converge to biologically plausible functionality in cortical circuits via task-optimization. This functionality has not been observed in CNN models of retinal circuits via task-optimization. We sought to observe this convergence in retinal circuits by designing a biologically inspired CNN model of a motion-detection retinal circuit and optimizing it to solve a motion-classification task. The learned weights and parameters indicated that the CNN converged to direction-sensitive ganglion and amacrine cells, cell types that have been observed in biology, and provided evidence that task-optimization is a fair method of building retinal models. The analysis used to understand the functionality of our CNN also indicates that biologically constrained deep learning models are easier to reason about their underlying mechanisms than traditional deep learning models. 

Abstract A new search for two-neutrino double-beta (2νββ) decay of¹³⁶Xe to theexcited state of¹³⁶Ba is performed with the full EXO-200 dataset. A deep learning-based convolutional neural network is used to discriminate signal from background events. Signal detection efficiency is increased relative to previous searches by EXO-200 by more than a factor of two. With the addition of the Phase II dataset taken with an upgraded detector, the median 90% confidence level half-life sensitivity of 2νββdecay to thestate of¹³⁶Ba isyr using a total¹³⁶Xe exposure of 234.1 kg yr. No statistically significant evidence for 2νββdecay to thestate is observed, leading to a lower limit ofyr at 90% confidence level, improved by 70% relative to the current world's best constraint. 

Abstract Generative Adversarial Networks trained on samples of simulated or actual events have been proposed as a way of generating large simulated datasets at a reduced computational cost. In this work, a novel approach to perform the simulation of photodetector signals from the time projection chamber of the EXO-200 experiment is demonstrated. The method is based on a Wasserstein Generative Adversarial Network — a deep learning technique allowing for implicit non-parametric estimation of the population distribution for a given set of objects. Our network is trained on real calibration data using raw scintillation waveforms as input. We find that it is able to produce high-quality simulated waveforms an order of magnitude faster than the traditional simulation approach and, importantly, generalize from the training sample and discern salient high-level features of the data. In particular, the network correctly deduces position dependency of scintillation light response in the detector and correctly recognizes dead photodetector channels. The network output is then integrated into the EXO-200 analysis framework to show that the standard EXO-200 reconstruction routine processes the simulated waveforms to produce energy distributions comparable to that of real waveforms. Finally, the remaining discrepancies and potential ways to improve the approach further are highlighted.