48670 research outputs found
Sort by
Active Simultaneous Localization and Mapping in Unstructured Environment with a Quadrotor
Get PDFAn agent performing Simultaneous Localization and Mapping (SLAM) constructs a map of the environment while estimating its location at the same time. SLAM algorithms primarily focus on finding the best estimate of the location of the agent, and by extension, that of the landmarks in the environment, given observations from sensors. These algorithms do not typically address how an agent should explore an unknown environment to build a map efficiently. This ability for active exploration is important for autonomous robots to work in unknown, unstructured environments such as forests or caves.
This paper proposes an active SLAM system that allows an agent to explore its surroundings, using visual-inertial data from an RGBD camera. We formalize this problem as taking actions that maximize the amount of information obtained from the scene. At each time step, a utility function that computes the incremental information gain is used to take actions. We conduct experiments using an Intel RealSense camera mounted on a custom-built quadrotor and show that we can explore indoor environments (while restricting the actions to choosing new viewpoints in SO(3) for practical reasons)
Developing a Methodology for the Conservation of a Tserkva: Examining the structural and Decorative Composite of the Church of the Exaltation of the Holy Cross, Drohobych, Ukraine
Get PDFIn January 2020, a multi-year project was initiated to preserve the Church of the Exaltation of the Holy Cross (CEHC), a historic wooden church in Western Ukraine. Building upon preliminary documentation, this thesis responds to the conservation challenge presented by the polychrome wall-paintings where the icons and motifs used in the religious liturgy were painted directly onto the structural timbers. This thesis proposes a methodology for their joint conservation informed by examination of the church’s heritage values, regional agents of deterioration, and conservation discourse around the larger typology. Through developed preservation principles and a phased timeline, the site-specific needs of the church are considered in order to aid resource managers in prioritizing future conservation work
Late-Childhood Foundational Cognitive Skills Predict Educational Outcomes Through Adolescence and Into Young Adulthood: Evidence from Ethiopia and Peru
Get PDFWe estimate the associations between a set of foundational cognitive skills (inhibitory control, working memory, long-term memory, and implicit learning) measured at age 12 and educational outcomes measured at ages 15 and 19-20 in Ethiopia and Peru (the Young Lives study). The estimates adjust for a rich set of lagged controls and include measurements of children’s general abilities. For a subset of the outcomes, we exploit within-household variation. Working memory and long-term memory are consistently and positively associated with subsequent domain-specific cognitive achievement tests in both countries, university enrolment in Peru (working memory) and lower secondary-school completion in Ethiopia (long-term memory). Inhibitory control predicts subsequent math-test scores in both countries, and grade attainment in Ethiopia. These results provide additional evidence to justify the importance of promoting investments in cognitive skills throughout childhood and adolescence, and these results potentially elucidate how investments in children impact their educational achievements
Explorations in Novel Photochemical Modes of Radical-Alkene Reactivity: My Piece of the π
No full textThe formation of Csp3-Csp3 bonds is arguably the most critical transformation in organic synthesis. For many years the Michael addition which utilizes highly reactivity organometallic carbon nucleophiles which are “poised to react” was used to achieve hydroalkylation of activated alkenes. Furthermore, this mode of reactivity also offered the potential to use other electrophiles to perform vicinal (1,2) alkene difunctionalization. Though effective in many contexts this reaction is often plagued by functional group incompatibilities, poor 1,2 vs 1,4 addition selectivity and unstable precursors. In contrast to organometallic carbanions, carbon radicals display much more selective reactivity, particularly with alkenes, and demonstrate much improved chemoselectivity. The Giese addition utilizes carbon radicals typically generated from alkyl halides via halogen atom abstraction to engage in hydroalkylation with activated alkenes. Unfortunately, the harsh conditions and reagents required for the radical chain mechanisms employed in Giese reactions restrict the transformation to alkene hydroalkylations with minimal functional group compatibility. However, several modes of photochemical catalysis can be used for radical generation which are significantly milder and more tolerance than typical radical initiation used in classical Giese addition protocols. These catalytic mechanisms offer the potential for further functionalization of radical addition intermediates opening the door a wide variety of alkene difunctionalizations with dramatically improved functional group compatibility compared to classical vicinal difunctionalizations. Herein are reported personal explorations in novel radical alkene reactions facilitated via photochemical modes of catalysis
The Role of Working Memory in Implementing Computational Elements of Visuo-Spatial Decision-Making
No full textDecision making is a dynamic process by which a person integrates sensory evidence with prior expectations to select an action to achieve a desired outcome. Often, this process requires working memory to hold and manipulate the relevant information. Working memory has several known limitations and models of implementation that have not been widely considered in the context of decision-making. For decisions regarding spatial stimuli, the dorsolateral prefrontal cortex (dlPFC) is likely to be essential. Activity in this area has been shown to relate to both decision making and working memory, and has led to concrete models of how spatial information may be stored by population activity. To investigate how this activity might be leveraged to implement several computational elements of decision making, we performed two related experiments. In the first, we had human participants perform a working memory task that required the reporting of a decision variable (average location) to determine how working memory limitations impacted decision precision. We also used models of working memory to predict and interpret what information was being actively maintained. We report not only the novel finding that decision variables held in working memory lose precision over time but also that the degree of precision loss depends on the strategy used to make the decision, which differed across participants and conditions. In our second study, we trained monkeys to perform an adaptive oculomotor delayed response task in which they had to integrate cue information with context to select the target most likely to be rewarded. The goal of this study was to investigate whether the tuning properties of working-memory related dlPFC neurons adjusted according to the statistics of the task, corresponding to adaptive behavior. We found that not all monkeys display adaptive behavior, but developed method for interrogating the dynamics of neural responses in those that do. This body of work contributes to our understanding of how working memory may implement the representation of a decision variable or prior information used to interpret incoming evidence. Such understanding may ultimately lead to more effective approaches to addressing disorders of maladaptive decisions, such as addiction and PTSD
Cosmology and Astrophysics from Small Scales
No full textCross-correlations between tracers of large-scale structure (LSS), such as galaxies, weak lensing, and thermodynamics of hot gas, provide powerful tests of the cosmological model. In this Ph.D. thesis, we develop analytical models of these tracers and apply them to compare measurements to theoretical predictions of the standard model of cosmology. The complicated non-linear interactions between various components of the Universe present a significant challenge to constraining cosmological or astrophysical models. We aim to maximize the information gained from current and future cosmological datasets in the presence of astrophysical and observational sources of uncertainty. In the first half of the thesis, we describe and validate a hybrid galaxy biasing model (non-linear mapping between dark matter and galaxies) aimed at analyzing the correlations between galaxy positions and weak lensing. We then apply this model to recent data from the Dark Energy Survey, leading to a significant gain in cosmological constraints. In the second half of the thesis, we carry out high significance measurements of cross-correlations between the pressure of hot gas and weak lensing (shear-y) and galaxy positions (galaxy-y). We constrain the evolution of the average thermal pressure of the Universe and find evidence for reduced pressure in low mass halos. Our results point to the effects of increased baryonic feedback (the impact of supernovae or active galactic nuclei on LSS). These results will help in understanding how baryonic feedback impacts galaxy formation and using the non-linear regime for cosmological analysis with future survey data
Asymptotic Properties of Disordered Systems
No full textThis thesis considers asymptotic behaviors of high-dimensional disordered systems, including Ising model and mean-field spin glass models. We first study the decay rate of correlations in the two-dimensional random field Ising model (RFIM). Second, we study the limit free energy of disordered systems. For RFIM, we are interested in the two-dimensional case where the external field is of i.i.d centered Gaussian variables. We show that under nonnegative temperature, the effect of boundary conditions on the magnetization in a finite box decays exponentially in the side length of the box. On the side of mean-field models, we use the Hamilton-Jacobi equation (HJE) approach, initiated by Jean-Christophe Mourrat, to characterize limiting free energy in many models from statistical inference problems and mean-field spin glass models. We now investigate infinite-dimensional models including many spin glass models and inference problems where the rank of the signal matrix increases as is sent to infinity. We give an intrinsic meaning to the Hamilton--Jacobi equation arising from mean-field spin glass models in the viscosity sense, and establish the corresponding well-posedness.This will shed more light on the mysterious Parisi formula as the limit of free energy in the Sherrington--Kirkpatrick model
The MLL4-LSD1 Epigenetic Axis Regulates Epidermal Differentiation and Ferroptosis: Implications for Skin Cancer and Disease
No full textEpigenetic mechanisms regulate chromatin structure and gene expression to direct organismal development and tissue homeostasis, disruption of which is pervasive in cancer. Epigenetic disruption can lead to the acquisition of each hallmark of cancer and almost half of all human cancers bear mutations in epigenetic regulators. The histone methyltransferase MLL4 (KMT2D) has been described as an essential gene in both humans and mice. In addition, it is one of the most commonly mutated genes in all of cancer biology with the highest frequency of mutation occurring in cutaneous squamous cell carcinomas (cSCC). The opposing demethylase LSD1 acts in a genetic and epigenetic axis with MLL4 and is highly overexpressed in various human cancers, which has resulted in the development of highly specific and potent LSD1 inhibitors. Despite this, how the MLL4-LSD1 epigenetic axis regulates normal epidermal homeostasis or becomes dysregulated to drive cSCC tumorigenesis is not well understood. Here, using transgenic mouse models, next-generation sequencing, and human skin models, we address this gap in knowledge. First, we identify a critical role for Mll4 in the promotion of epidermal differentiation and ferroptosis, a key mechanism of tumor suppression. Mice lacking Mll4, but not Mll3 (Kmt2c), display features of impaired differentiation and human precancerous neoplasms, all of which progress with age. Mll4 deficiency profoundly alters epidermal gene expression and uniquely rewires the expression of key genes and markers of ferroptosis (Alox12, Gpx4, Slc7a11). Beyond revealing a new mechanistic basis for Mll4-mediated tumor suppression, our data uncover a potentially much broader and general role for ferroptosis in the process of differentiation and skin homeostasis. In addition to these findings, we also show that LSD1 directly represses master epithelial transcription factors that promote differentiation. LSD1 inhibitors block both LSD1 binding to chromatin and its catalytic activity, driving significant increases in H3K4 methylation and gene transcription of these fate-determining transcription factors. This leads to both premature epidermal differentiation and the repression of features squamous cell carcinoma. Together, this work reveals critical functions for the MLL4-LSD1 epigenetic axis during epidermal differentiation, ferroptosis, and tumorigenesis and offer new therapeutic strategies to be tested for the treatment of cSCC
Penn Library\u27s LJS 395 - Manuscript pastedowns from De proprietatibus rerum (Video Orientation)
No full texthttps://repository.upenn.edu/sims_video/1149/thumbnail.jp
Penn Library\u27s Ms. Codex 1673 - Key of Solomon (Video Orientation)
No full texthttps://repository.upenn.edu/sims_video/1152/thumbnail.jp