Search for: All records

Abstract There is a growing demand for low-power, autonomously learning artificial intelligence (AI) systems that can be applied at the edge and rapidly adapt to the specific situation at deployment site. However, current AI models struggle in such scenarios, often requiring extensive fine-tuning, computational resources, and data. In contrast, humans can effortlessly adjust to new tasks by transferring knowledge from related ones. The concept of learning-to-learn (L2L) mimics this process and enables AI models to rapidly adapt with only little computational effort and data. In-memory computing neuromorphic hardware (NMHW) is inspired by the brain’s operating principles and mimics its physical co-location of memory and compute. In this work, we pair L2L with in-memory computing NMHW based on phase-change memory devices to build efficient AI models that can rapidly adapt to new tasks. We demonstrate the versatility of our approach in two scenarios: a convolutional neural network performing image classification and a biologically-inspired spiking neural network generating motor commands for a real robotic arm. Both models rapidly learn with few parameter updates. Deployed on the NMHW, they perform on-par with their software equivalents. Moreover, meta-training of these models can be performed in software with high-precision, alleviating the need for accurate hardware models. 

Abstract Ocean weather comprises vortical and straining mesoscale motions, which play fundamentally different roles in the ocean circulation and climate system. Vorticity determines the movement of major ocean currents and gyres. Strain contributes to frontogenesis and the deformation of water masses, driving much of the mixing and vertical transport in the upper ocean. While recent studies have shown that interactions with the atmosphere damp the ocean’s mesoscale vorticesO(100) km in size, the effect of winds on straining motions remains unexplored. Here, we derive a theory for wind work on the ocean’s vorticity and strain. Using satellite and model data, we discover that wind damps strain and vorticity at an equal rate globally, and unveil striking asymmetries based on their polarity. Subtropical winds damp oceanic cyclones and energize anticyclones outside strong current regions, while subpolar winds have the opposite effect. A similar pattern emerges for oceanic strain, where subtropical convergent flow is damped along the west-equatorward east-poleward direction and energized along the east-equatorward west-poleward direction. These findings reveal energy pathways through which the atmosphere shapes ocean weather. 

Abstract Many thermobarometric methods applied to granitic composition rocks that crystallized at < 5 kbar yield temperature estimates ~ 50 to 100 °C lower than the widely used haplogranite water-saturated solidus. To address thermobarometric discrepancies, we investigated a shallow-level granitic pluton that is not enriched in fluxing elements such as Li, B, P, or F. The Rito del Medio pluton (RDMP) near Questa, New Mexico contains numerous minerals and inclusions suitable for thermobarometric estimates, and it has abundant miarolitic cavities that represent the final crystallization stages. Occurrences of coexisting melt and fluid inclusions show that groundmass minerals crystallized from a water-saturated magma. After groundmass crystallization, the pluton transitioned to a fluid-dominated system manifested by the crystallization of freestanding minerals contained in the miarolitic cavities. The granite contains mica, feldspar, quartz, primary fluid inclusions in quartz, and accessory minerals including garnet with quartz inclusions. We used minerals and their inclusions in both paragenetic contexts to track changes in P, T, and mineral and fluid compositions that accompanied the magmatic-to-hydrothermal transition. Resultant univariant curves from thermobarometry for the groundmass minerals converge at ~ 1.9 to 2.1 kbar and ~ 590 to 625 °C indicating final magmatic crystallization. To address discrepancies between thermobarometric results and the haplogranite solidus, we performed crystallization experiments at 2 kbar, which show that RDMP compositions magmas complete crystallization at temperatures ~ 620–625 °C. Univariant curves for thermobarometric approaches applied to the RDMP miarolitic cavity minerals converge at ~ 1.5 to 2.1 kbar and ~ 500 to 525 °C defining the transition to hydrothermal crystallization conditions. 

\We show that braid varieties for any complex simple algebraic group G are cluster varieties. This includes open Richardson varieties inside the flag variety G/B. 

In response to COVID-19, the CDC issued hygiene, protective equipment, and physical distancing guidelines to reduce virus transmission. Adherence was crucial for public health, particularly in the earliest stage of COVID-19, before effective treatments emerged. Still, there was wide variation in willingness and/or ability to follow the recommendations. One group that might be expected to flout rules and take risks under normal circumstances is adolescents. This developmental stage predisposes one to push boundaries and seek the company of peers. Adolescents with a history of lawbreaking might be even more inclined to disregard public health guidelines due to experiential and dispositional factors. We employed a longitudinal study launched prior to the pandemic to identify which pre-pandemic factors predict adolescents’ adherence to—or disregard for—public health guidelines during a crisis. The sample (N = 75, 30% justice-involved) came from predominantly minoritized communities in a southwestern U.S. city. Data were collected in three waves over one year. Analyses tested whether adherence varied by time period, local infection trajectories, justice involvement, pre-pandemic mental health, risk-taking, and rule orientation. Results revealed that adherence declined over time and was generally lower among justice-involved adolescents. In addition, justice-involved adolescents with higher depressive symptoms displayed lower adherence, whereas those reporting higher anxiety symptoms displayed higher adherence. Understanding these factors is crucial for developing strategies to promote adherence to public health guidelines among adolescents during public health emergencies. 

Non-rigid spatial thinking, or mental transformations where the distance between two points in an object changes (e.g., folding, breaking, bending), is required for many STEM fields but remains critically understudied. We developed and tested a non-rigid, ductile spatial skill measure based on reasoning about knots with 279 US adults (M = 30.90, SD 5.47 years; 76% White; 48% women). The resultant 54-item measure had good reliability (α = .88). Next, 147 US adults (M = 20.65, SD 2.80 years; 48% White; 56% women) completed existing spatial skills measures, the knot reasoning measure, a verbal skill measure, and surveys of current and childhood spatial activities. Knot reasoning performance was significantly, positively correlated with existing measures of spatial skill. Mental rotation and paper folding, but not bending, predicted knot reasoning task performance. We replicated work showing that men performed better than women on mental rotation and unexpectedly found that men also outperformed women on paper folding and knot reasoning, but not bending, tasks. Using structural equation modeling, we found several significant mediation effects. Men who reported less masculine-stereotyped spatial activity engagement had higher performance on the mental rotation and knot reasoning tasks. Women who reported greater engagement in feminine-stereotyped spatial activities had higher paper folding and backwards knot reasoning performance. Spatial skills did not differ among math-intensive STEM, non-math-intensive STEM, and non-STEM majors. The studies introduce a reliable measure of non-rigid, ductile string transformations and provide initial evidence of the role of gender and gendered spatial activities on non-rigid spatial skills. 

Abstract The learning and recognition of object features from unregulated input has been a longstanding challenge for artificial intelligence systems. Brains, on the other hand, are adept at learning stable sensory representations given noisy observations, a capacity mediated by a cascade of signal conditioning steps informed by domain knowledge. The olfactory system, in particular, solves a source separation and denoising problem compounded by concentration variability, environmental interference, and unpredictably correlated sensor affinities using a plastic network that requires statistically well-behaved input. We present a data-blind neuromorphic signal conditioning strategy, based on the biological system architecture, that normalizes and quantizes analog data into spike-phase representations, thereby transforming uncontrolled sensory input into a regular form with minimal information loss. Normalized input is delivered to a column of spiking principal neurons via heterogeneous synaptic weights; this gain diversification strategy regularizes neuronal utilization, yoking total activity to the network’s operating range and rendering internal representations robust to uncontrolled open-set stimulus variance. To dynamically optimize resource utilization while balancing activity regularization and resolution, we supplement this mechanism with a data-aware calibration strategy in which the range and density of the quantization weights adapt to accumulated input statistics.