12023 research outputs found
Sort by
Earthquakes and the New Paradigm of Diluted Cores in Gas Giant Planets
Get PDFIn this thesis, I present results on two distinct topics within geophysics: earthquake mechanics and the core of gas giant planets. A common element connecting this work is the similar research approach that I use to address each topic. Each chapter in this thesis attempts to provide a simple physical understanding on the fundamental aspects relevant to the system in question. Further, I use numerical models to expand my arguments in some cases, while in others I build up my case with mathematical modeling only.
Chapters II-IV focus on the gravitational field of Jupiter and connect radio science observations from NASA's Juno mission to the structure of Jupiter's dilute core. In Chapter II, I use dynamical tides to interpret a nonhydrostatic component in Jupiter's degree-2 tidal response -- represented by the Love number k₂ -- observed by Juno at the mid-mission perijove (PJ) 17. The results presented here show how the Coriolis acceleration contributes with a dynamical effect to Jupiter's tidal response, providing a satisfactory fit to Juno's observed k₂. From these results, I conclude that Juno obtained the first unambiguous detection of the gravitational effect of dynamical tides in a gas giant planet.
In Chapter III, I build a perturbation theory to show that the high-degree tidal gravitational field of Jupiter is dominated by spherical harmonic coupling promoted by Jupiter's oblate figure as forced by the centrifugal effect. Based on this novel understanding of Jupiter's high-degree tidal gravitational field, I establish that Juno observed a 7σ nonhydrostatic component in k₄₂ at mid-mission.
In Chapter IV, I invoke a core-orbital resonance between internal gravity waves trapped in Jupiter's dilute core and the orbital motion of Io to explain the 7σ nonhydrostatic component in the high-degree tidal response of Jupiter as observed by Juno at mid-mission -- namely the Love number k₄₂. These results suggest that an extended dilute core in Jupiter (r ≳ 0.7RJup) reconciles the k₄₂ nonhydrostatic component. This explanation of Juno's observation requires two ingredients: a dilute core in Jupiter that becomes smoother or shrinks over geological time, alongside with a high amount of dissipation provided by resonantly excited internal gravity waves.
In Chapter V, I connect observations of earthquake modes of propagation to the damaged rock often found around tectonic fault zones. Previous work showed that pulse-like rupture -- a propagation mode where slip propagates as a narrow pulse -- can be induced by the dynamic effect of seismic waves reflected at the boundary of a cavity formed by the damaged material in fault zones. My main result shows that pulses are easier to produce than previously thought; pulses can appear in a highly damaged fault zone even in the absence of reflected seismic waves. In addition, these results provide a new explanation for back-propagating rupture fronts recently observed during large earthquakes and the rapid-tremor-reversal slip patterns observed in Cascadia and Japan.
In summary, the results contained in these four chapters advance our knowledge in fundamental problems related to geophysics. In relation to gas giant planets, my results include the development of a novel technique to reveal the structure of Jupiter's core using spacecraft observations of the tidal gravitational field. In relation to earthquakes, my results connect earthquake ruptures to observable fault zone properties.</p
Autonomous Temporal Understanding and State Estimation during Robot-Assisted Surgery
Get PDFRobot-Assisted Surgery (RAS) has become increasingly important in modern surgical practice for its many benefits and advantages for both the patient and the healthcare professionals, as compared to traditional open surgeries and minimally invasive surgeries such as laparoscopy. Artificial intelligence applications during RAS and post-operative analysis can provide various surgeon-assisting functionalities and could potentially achieve a better surgery outcome. These applications, ranging from providing surgeons with advisory information during RAS and post-operative analysis to virtual fixture and supervised autonomous surgical tasks, share a necessary prerequisite of a comprehensive understanding of the current surgical scene. This understanding should include the knowledge of the current surgical task being performed, the surgeon's actions and gestures, the state of the patient, etc. Currently, there is yet to be a unified effort to achieve the autonomous temporal understanding and perception of an RAS at the high accuracy and efficiency required in the highly safety-critical field of medicine.
This thesis develops novel modeling methodologies and deep learning-based models for the autonomous perception and temporal segmentation of the current surgical scene during an RAS. An RAS procedure is modeled as a hierarchical system consisting of discrete surgical states at multiple levels of temporal granularity. These surgical states take the form of surgical tasks, operational steps, fine-grained surgical actions, etc. A broad range of computational experiments were performed to develop methods that achieve an accurate, robust, and efficient estimation of these surgical states. Multiple novel deep learning-based models for feature extraction, noise elimination, and efficient training were proposed and tested. This thesis also shows the significant benefits of incorporating multiple types of data streams recorded by the surgical robotic system to a more accurate surgical state estimation effort.
Two new RAS datasets that contains real-world RAS procedures and diverse experimental settings were collected and annotated--filling a gap in the data sets available for the development and testing of of robust surgical state estimation models. The performance and robustness of models in this thesis work were showcased with these highly complex and dynamic real-world RAS datasets and compared against state-of-the-art methods. A significant model performance improvement was observed in both surgical state estimation accuracy and efficiency. The modeling methodologies and deep learning-based models developed in this work have diverse potential applications to the development of a next-generation surgical robotic systems.</p
Optimization of Distribution Power Networks: from Single-Phase to Multi-Phase
Get PDFDistributed energy resources play an important role in today's distribution power system. The Optimal Power Flow (OPF) problem is fundamental in power systems as many important applications such as economic dispatch, battery displacement, unit commitment, and voltage control can be formulated as an OPF. A paradoxical observation is the problem's complexity in theory but simplicity in practice. On the one hand, the problem is well known to be non-convex and NP-hard, so it is likely that no simple algorithms can solve all problem instances efficiently. On the other hand, there are many known algorithms which perform extremely well in practice for both standard test cases and real-world systems. This thesis attempts to reconcile this seeming contradiction.
Specifically, this thesis focuses on two types of properties that may underlie the simplicity in practice of OPF problems. The first property is the exactness of relaxations, meaning that one can find a convex relaxation of the original non-convex problem such that the two problems share the same optimal solution. This property would allow us to convexify the non-convex problem without altering the optimal solution and cost. The second property is that all locally optimal solutions of the non-convex problem are also globally optimal. This property allows us to apply local algorithms such as gradient descent without being trapped at some spurious local optima. We focus on distribution systems with radial networks (i.e., the underlying graphs are trees). We consider both single-phase models and unbalanced multi-phase models, since most real-world distribution systems are multi-phase unbalanced, and distributed energy resources (DERs) can be connected in either Wye or Delta configurations.
The main results of this thesis are two-fold. In the first half, we propose a class of sufficient conditions for a non-convex problem to simultaneously have exact relaxation and no spurious local optima. Then we apply the result to single-phase system and conclude that if all buses have no injection lowerbounds, then both properties (exactness and global optimality) can be achieved. While the same condition is already known to be sufficient for exactness, our work is the first to extend it to global optimality. In the second half, we focus on the exactness property for multi-phase systems. For systems without Delta connections, the exactness can be guaranteed if 1) the binding constraints are sparse in the network at optimality; or 2) all nodal prices fall within a narrow range. Using the DC model as an approximation, we further analyze the OPF sensitivity and explain why nodal prices tend to be close to each other. In the presence of Delta connections, we conclude that the inexactness can be resolved by either postprocessing an optimal solution, or adding a new regularization term in the cost function. Both methods achieve global optimality for IEEE standard test cases.</p
Imaging Neuropeptide Release and Localization with Genetically Engineered Reporters
Get PDFNeuropeptides are a class of neural signaling molecules that play a pivotal role in brain function and human health through neuromodulatory influences. There are over 100 types of neuropeptides identified and characterized, yet genomic analysis suggests that it is only the tip of the iceberg, with extra hundreds of putative neuropeptides awaiting further investigation. Neuropeptides collectively regulate a variety of developmental, physiological, and behavioral functions. While each neuropeptide is idiosyncratic in regard to its molecular structure, chemical properties, and anatomical distribution, they impinge on the nervous system in a similar fashion.
Surprisingly, despite their fundamental importance, techniques for measuring the localization, expression and release of neuropeptides, at large scale and with high spatiotemporal resolution, have lagged far behind. Microdialysis and fast-scanning cyclic voltammetry are useful primarily for measuring "volume transmission," but are invasive, and have poor spatial resolution and limited general applicability. FP-tagged vesicle reporters are mainly tested and used in limited cell types. Little is characterized about their functional universality and specificity. GPCR-based sensors are designed to visualize the binding, instead of expression and release, of a neuropeptide.
Therefore, I aim to develop new methods for visualizing, detecting, and inhibiting NP expression and release in vivo. The long-term goal is to apply these methods to understanding the dynamics of neuromodulation of specific, behaviorally relevant neural circuits, and to providing a dynamic, high-resolution view of chemical modulation of circuit function.
In Chapter 2, I will describe the design, screening, and proof-of-concept validation of novel genetically engineered neuropeptide release reporters (NPRR) in Drosophila. I further demonstrated the idiosyncrasy of neuropeptide release dynamics, as well as cell-type specific release properties of a neuropeptide. In Chapter 3, I conceived and constructed a neuropeptide imaging platform that exploits the discoveries and strategies from Drosophila NPRRs. Besides a series of redesign of mammalian NPRRs, a collection of sister reporters to visualize localization and expression (Neuropeptide Localization and Expression Reporter, NPLER) were built in parallel. I also established a prototypical pipeline to systematically screen for appropriate cell lines for the purpose of NPRR/NPLER applications.
Malfunctioning of neuropeptide pathways can potentially result in a variety of mental illnesses triggered by stress, and metabolic disorders including obesity. Drugs targeting neuropeptide signaling have received heavy investment, but most have failed in the clinical trials. We therefore propose alternative strategies to target the processing/release of the neuropeptide from neurons, rather than blocking its receptor. In Chapter 4, I describe the ongoing process of adapting modern biotechnologies to the imaging platform to explore novel therapeutic strategies for neuropeptide- relevant disorders and abnormalities.
The Appendix includes a serendipitous finding from our attempt to generalize NPRR to Caenorhabditis elegans.</p
Spin-Phonon Interactions and Spin Decoherence from First Principles
Get PDFDeveloping a microscopic understanding of spin decoherence is essential to advancing quantum technologies. Electron spin decoherence due to atomic vibrations (phonons) plays a special role as it sets an intrinsic limit to the performance of spin-based quantum devices. Two main sources of phonon-induced spin decoherence, the Elliott-Yafet (EY) and Dyakonov-Perel (DP) mechanisms, have distinct physical origins and theoretical treatments. First-principles calculations of electron-phonon (e-ph) interactions combined with many-body perturbation theory are promising to study phonon-induced spin decoherence. However, predicting the spin response in materials remains an open challenge; methods for quantifying spin-dependent e-ph interactions in materials, as well as a linear response framework for spins in the presence of e-ph interaction is missing. In this thesis, we provide a first-principles framework for computing the relativistic spin-dependent electron-phonon interactions. We develop a formalism that unifies the modeling of EY and DP spin decoherence, and provide a rigorous many-body perturbation theory for obtaining the spin-spin correlation function including the vertex corrections due to e-ph interactions. We compute the phonon-dressed vertex of the spin-spin correlation function with a treatment analogous to the calculation of the anomalous electron magnetic moment in QED. We find that the vertex correction provides a giant renormalization of the electron spin dynamics in solids, greater by many orders of magnitude than the corresponding correction from photons in vacuum. We further identify the long-range quadrupole e-ph interaction in materials, and demonstrate its importance in the description of phonon-induced spin decoherence. We show first-principle calculations of spin-dependent e-ph interactions in correlated electron systems, using the framework of Hubbard-corrected density functional theory. Lastly, we provide technical details in the implementation of ab-initio e-ph interaction in PERTURBO, a software package for first-principles calculations of charge transport, spin dynamics, and ultrafast carrier dynamics in materials. In summary, the thesis demonstrates a general approach for quantitative analysis of spin decoherence in materials, advancing the quest for spin-based quantum technologies.</p
An Efficient Environment Encoding for Trajectory Feasibility and Cost Predictions
Get PDFAn efficient method to encoding a robot’s surrounding is important for the safety and scalability of path planning problems. I present a method for encoding occupied spaces in an environment via parametric geometry (e.g. circles). This encoding is utilized to train a cost and feasibility predicting neural network for path planning between two states. The circle encoding method uses either bounding circles or set cover to provide a lower dimension input of environments for neural network. The lower input dimension of the models allows for a smaller and faster model of similar performance. Feasibility and cost prediction neural networks enable lazy planning for trajectory calculation. While there are no safety-guarantees, the model I present can act as an heuristic to identify the best paths, amortizing a robot’s onboard computation. To train a cost and feasibility predicting model, I first generate multiple environments and calculate encoding circles for obstacle-occupied spaces. For each environment, I use planners to calculate multiple trajectories and store their feasibility and cost. I then train the feasibility and cost prediction models from this data . The feasibility model has 90% accuracy on train and test environments. The circle encoding method was compared with occupancy grid encoding method. When encoding methods were compared to predict occupied space, models with circle encoding generalized to unseen environments better than occupancy grid input models. The method of circle encoding I introduce can be utilized in other learning problems that use environment as input. The feasibility and cost predictions done in this work can be applied to bias tree growth or to other global planning methods
Nucleic Acid Measurements for Antibiotic Susceptibility Testing and Early Detection of SARS-CoV-2
Get PDFNucleic-acid-amplification tests (NAATs) are widely used in microbial detection both in environmental characterization and human diagnostics. NAATs offer highly sensitive and specific detection of target molecules among the noise of complex samples. This thesis covers two important applications of nucleic-acid quantification techniques in human clinical samples. First, I co-developed a new phenotypic antibiotic susceptibility test that uses species-specific DNA detection to detect bacterial cell-wall damage following incubation with beta-lactam antibiotics. Second, I helped compile a longitudinal dataset of SARS-CoV-2 viral loads during a community-based COVID-19 study run by the Ismagilov Lab through October 2020 – April 2021 in the greater Los Angeles County area, USA. Sensitive and specific nucleic-acid tests allowed for robust detection of pathogenic microbes in both these applications. Designing and implementing NAATs for these applications required consideration of biological constraints of the microorganisms, molecular stability over the time of quantification, and the practical constraints of acquiring and transporting samples. Continued innovation of NAAT technologies will be critical to contain present and future pandemics and empower medical professionals with data to inform treatment options.</p
Model Parameterization and Model Selection in Geophysical Inverse Problems. Designing Inverse Problems that Respect a priori Geophysical Knowledge
Get PDFThe vast majority of the Earth system is inaccessible to direct observation. Consequently, the structure and dynamics of the Earth can only be determined indirectly, via geophysical sensing. These methods have the mathematical form of an inverse problem, in which the data and the unknowns are linked by a physical process, such as seismic wave propagation. From the possibly noisy data, we have indirect access to the unknowns. The vast majority of geophysical inverse problems are ill-posed, and require the provision of a priori knowledge to stabilize the solution. This thesis investigates methods for designing inverse problems to better take advantage of geophysical or geological constraints, to allow better resolution or more interpretability of the solutions. Four major themes are investigated: In Chapter 2, we study the collection of a novel dataset of Rayleigh wave horizontal-to-vertical ratios to provide stronger constraints on upper-crustal structure in Southern California. In Chapters 3 and 4, we develop a method for wavefield-reconstruction of sparse seismic data, including heterogeneous networks consisting of both displacement and strain instruments. This method amounts to an inversion in data-space, and promises to unlock the potential of wavefield based methods for complex datasets. In Chapters 5 and 6, we investigate a new structural parameterization based on a combination of Gaussian processes and the level-set method, that better models discontinuous geological features such as sedimentary basins. We test our method on a variety of synthetic and real datasets, culminating in a detailed study of the northeastern Los Angeles basin, which we found to be significantly deeper and steeper than in previous models. Finally, we develop a method of model selection for noisy historical datasets, which we investigate using the case study of correcting Oldham's data misinterpretation in the 1906 paper that "discovered" Earth's core.</p
The Synthesis of Diverse Families of Organic Compounds via Nickel-Catalyzed Nucleophilic Substitution Reactions
Get PDFTransition metal-catalyzed cross-coupling has provided an exceptionally powerful approach to carbon–carbon bond formation, allowing chemists to solve a number of important problems in organic synthesis. However, by the early 2000s, its application to the formation of alkyl–alkyl bonds had been limited by the slow oxidative addition of palladium catalysts toward alkyl halides and the tendency of transition-metal-alkyls to undergo β-hydride elimination. Since then, complexes based on nickel, an earth-abundant metal, have emerged as efficient catalysts for the nucleophilic substitution of alkyl electrophiles. The propensity for nickel to access a range of oxidation states allows it to react via one-electron pathways to generate radical intermediates, opening the door to couplings of sterically-hindered electrophiles and offering a ready mechanism for enantioconvergence.
Our group has applied nickel catalysts to substitution reactions of activated and unactivated 2o and 3o alkyl electrophiles by carbon– as well as by heteroatom-based nucleophiles, including a number of enantioconvergent processes. However, given the enormous range of conceivable coupling partners, many interesting challenges have yet to be addressed. The application of nickel-catalyzed substitution reactions to the synthesis of diverse families of compounds, particularly those with frequent uses in organic synthesis and pharmaceutical science, is described in this thesis. While reaction development is the primary focus of this work, a variety of synthetic applications and mechanistic investigations are also detailed within.
Chapter 2 describes two methods for the catalytic enantioconvergent synthesis of amines, which involve the coupling of an alkylzinc reagent with a racemic electrophile (specifically, an α-phthalimido alkyl chloride and an N-hydroxyphthalimide ester of a protected α-amino acid). A one-pot variant of this transformation is possible, enabling the efficient enantioselective synthesis of a range of interesting target molecules. Several mechanistic insights are also detailed.
Chapter 3 outlines the nickel-catalyzed alkylation of racemic α-haloglycine derivatives, a class of electrophile previously unemployed in metal-catalyzed asymmetric cross-coupling reactions, with alkylzinc reagents to generate protected unnatural α-amino acids. This method is applied to the generation of several enantioenriched unnatural α-amino acids that have previously been shown to serve as useful intermediates in the synthesis of bioactive compounds.
Chapter 4 details the development of a nickel-catalyzed cross-coupling for the asymmetric synthesis of protected thiols. The synthesis of an N-hydroxyphthalimide ester containing a geminal thioester (a previously unreported class of molecule with no applications to cross-coupling) is described, along with its reactivity toward alkylzinc reagents and other classes of organometallic nucleophiles.
Chapter 5 examines the ability of nickel to catalyze the nucleophilic fluorination of unactivated alkyl halides, a transformation whose application to the synthesis of alkyl fluorides has been impeded by the low nucleophilicity and high basicity of fluoride. The reactivities of unactivated 1o, 2o, and 3o alkyl bromides, as well as several preliminary mechanistic investigations, are presented.</p
The Neural Mechanisms of Value Construction
Get PDFResearch in decision neuroscience has characterized how the brain makes decisions by assessing the expected utility of each option in an abstract value space that affords the ability to compare dissimilar options. Experiments at multiple levels of analysis in multiple species have localized the ventromedial prefrontal cortex (vmPFC) and nearby orbitofrontal cortex (OFC) as the main nexus where this abstract value space is represented. However, much less is known about how this value code is constructed by the brain in the first place. By using a combination of behavioral modeling and cutting edge tools to analyze functional magnetic resonance imaging (fMRI) data, the work of this thesis proposes that the brain decomposes stimuli into their constituent attributes and integrates across them to construct value. These stimulus features embody appetitive or aversive properties that are either learned from experience or evaluated online by comparing them to previously experienced stimuli with similar features. Stimulus features are processed by cortical areas specialized for the perception of a particular stimulus type and then integrated into a value signal in vmPFC/OFC.
The project presented in Chapter 2 examines how food items are evaluated by their constituent attributes, namely their nutrient makeup. A linear attribute integration model succinctly captures how subjective values can be computed from a weighted combination of the constituent nutritive attributes of the food. Multivariate analysis methods revealed that these nutrient attributes are represented in the lateral OFC, while food value is encoded both in medial and lateral OFC. Connectivity between lateral and medial OFC allows this nutrient attribute information to be integrated into a value representation in medial OFC.
In Chapter 3, I show that this value construction process can operate over higher-level abstractions when the context requires bundles of items to be valued, rather than isolated items. When valuing bundles of items, the constituent items themselves become the features, and their values are integrated with a subadditive function to construct the value of the bundle. Multiple subregions of PFC including but not limited to vmPFC compute the value of a bundle with the same value code used to evaluate individual items, suggesting that these general value regions contextually adapt within this hierarchy. When valuing bundles and single items in interleaved trials, the value code rapidly switches between levels in this hierarchy by normalizing to the distribution of values in the current context rather than representing all options on an absolute scale.
Although the attribute integration model of value construction characterizes human behavior on simple decision-making tasks, it is unclear how it can scale up to environments of real-world complexity. Taking inspiration from modern advances in artificial intelligence, and deep reinforcement learning in particular, in Chapter 4 I outline how connectionist models generalize the attribute integration model to naturalistic tasks by decomposing sensory input into a high dimensional set of nonlinear features that are encoded with hierarchical and distributed processing. Participants freely played Atari video games during fMRI scanning, and a deep reinforcement learning algorithm trained on the games was used as an end-to-end model for how humans evaluate actions in these high-dimensional tasks. The features represented in the intermediate layers of the artificial neural network were found to also be encoded in a distributed fashion throughout the cortex, specifically in the dorsal visual stream and posterior parietal cortex. These features emerge from nonlinear transformations of the sensory input that connect perception to action and reward. In contrast to the stimulus attributes used to evaluate the stimuli presented in the preceding chapters, these features become highly complex and inscrutable as they are driven by the statistical properties of high-dimensional data. However, they do not solely reflect a set of features that can be identified by applying common dimensionality reduction techniques to the input, as task-irrelevant sensory features are stripped away and task-relevant high-level features are magnified.</p