Search

This study captured middle and high school teachers’ perceptions of what they learned from professional development (PD) 3–4 years after participating in one of three National Science Foundation funded year-long PD projects. We surveyed 66 teachers from three different PD projects on the types of content, pedagogy, and resources that they remembered learning and continue to use when teaching mathematics. Results indicate that teachers remember and use many aspects from their PD experiences 3–4 years down the road. Most residual learnings from PD also appear to be highly aligned with the goals and intentions of the PD developers and researchers and may be related to the kind of PD design on the adaptive-specified continuum.

High-performance batteries are poised for electrification of vehicles and therefore mitigate greenhouse gas emissions, which, in turn, promote a sustainable future. However, the design of optimized batteries is challenging due to the nonlinear governing physics and electrochemistry. Recent advancements have demonstrated the potential of deep learning techniques in efficiently designing batteries, particularly in optimizing electrodes and electrolytes. This review provides comprehensive concepts and principles of deep learning and its application in solving battery-related electrochemical problems, which bridges the gap between artificial intelligence and electrochemistry. We also examine the potential challenges and opportunities associated with different deep learning approaches, tailoring them to specific battery requirements. Ultimately, we aim to inspire future advancements in both fundamental scientific understanding and practical engineering in the field of battery technology. Furthermore, we highlight the potential challenges and opportunities for different deep learning methods according to the specific battery demand to inspire future advancement in fundamental science and practical engineering.

Abstract The urban heat island effect causes increased heat stress in urban areas. Cool roofs and urban greening have been promoted as mitigation strategies to reduce this effect. However, evaluating their efficacy remains a challenge, as potential temperature reductions depend on local characteristics. Existing methods to characterize their efficacy, such as computational fluid dynamics and urban canopy models, are computationally burdensome and require a high degree of expertise to employ. We propose a data-driven approach to overcome these hurdles, inspired by recent innovations in spatial causal inference. This approach allows for estimates of hypothetical interventions to reduce the urban heat island effect. We demonstrate this approach by modeling evening temperature in Durham, North Carolina, using readily retrieved air temperature, land cover, and satellite data. Hypothetical interventions such as lining streets with trees, cool roofs, and changing parking lots to green space are estimated to decrease evening temperatures by a maximum of 0.7–0.9 $$^{\circ } \hbox {C}$$ ∘ C , with reduced effects on temperature as a function of distance from the intervention. Because of the ease of data access, this approach may be applied to other cities in the U.S. to help them come up with city-specific solutions for reducing urban heat stress.

Abstract Dielectric metasurfaces, composed of planar arrays of subwavelength dielectric structures that collectively mimic the operation of conventional bulk optical elements, have revolutionized the field of optics by their potential in constructing high-efficiency and multi-functional optoelectronic systems on chip. The performance of a dielectric metasurface is largely determined by its constituent material, which is highly desired to have a high refractive index, low optical loss and wide bandgap, and at the same time, be fabrication friendly. Here, we present a new material platform based on tantalum pentoxide (Ta2O5) for implementing high-performance dielectric metasurface optics over the ultraviolet and visible spectral region. This wide-bandgap dielectric, exhibiting a high refractive index exceeding 2.1 and negligible extinction coefficient across a broad spectrum, can be easily deposited over large areas with good quality using straightforward physical vapor deposition, and patterned into high-aspect-ratio subwavelength nanostructures through commonly-available fluorine-gas-based reactive ion etching. We implement a series of high-efficiency ultraviolet and visible metasurfaces with representative light-field modulation functionalities including polarization-independent high-numerical-aperture lensing, spin-selective hologram projection, and vivid structural color generation, and the devices exhibit operational efficiencies up to 80%. Our work overcomes limitations faced by scalability of commonly-employed metasurface dielectrics and their operation into the visible and ultraviolet spectral range, and provides a novel route towards realization of high-performance, robust and foundry-manufacturable metasurface optics.

Convolutional Neural Networks (CNNs) are a promising technique to predict highly localized fine particulate matter (i.e., PM2.5 levels) based on high-resolution satellite imagery. Unfortunately, CNNs typically require large amounts of supervised data to perform well, whereas this application generally has lots of unsupervised data (all satellite imagery) and relatively sparse supervised data (measurements from ground sensors). Previous work used transfer learning from another visual task to initialize the CNN weights; however, we hypothesize that standard transfer learning strategies would bias the CNN to focus on irrelevant details of the image for our applications. Instead, we develop a novel framework called Spatiotemporal Contrastive Learning (SCL) to pre-train the CNN. We test both regular contrastive learning and SCL on predicting PM2.5 levels from satellite images in two different cities, Delhi and Beijing, and compare to CNNs with parameters initialized randomly and by transfer learning. Our results show that regular contrastive learning and our SCL frameworks both manage to better capture spatial variation of ground-level PM2.5 concentrations compared to traditional initialization schemes, and that this performance gap increases as the number of ground sensors decreases, implying that the approach will be even more valuable in cities with fewer ground sensors. Our work demonstrates that contrastive learning is a powerful pre-training technique to build better spatial maps of PM2.5, and can be broadly applied in related situations.

The plant genus Oenothera has played an important role in the study of plant evolution of genomes and plant defense and reproduction. Here, we build on the 1kp transcriptomic dataset by creating 44 new transcriptomes and analyzing a total of 63 transcriptomes to present a large-scale comparative study across 29 Oenothera species. Our dataset included 30.4 million reads per individual and 2.3 million transcripts on average. We used this transcriptome resource to examine genome-wide evolutionary patterns and functional diversification by searching for orthologous genes and performed gene family evolution analysis. We found wide heterogeneity in gene family evolution across the genus, with section Oenothera exhibiting the most pronounced evolutionary changes. Overall, more significant gene family expansions occurred than contractions. We also analyzed the molecular evolution of phenolic metabolism by retrieving proteins annotated for phenolic enzymatic complexes. We identified 1,568 phenolic genes arranged into 83 multigene families that varied widely across the genus. All taxa experienced rapid phenolic evolution (fast rate of genomic turnover) involving 33 gene families, which exhibited large expansions, gaining about 2-fold more genes than they lost. Upstream enzymes phenylalanine ammonia-lyase (PAL) and 4-coumaroyl: CoA ligase (4CL) accounted for most of the significant expansions and contractions. Our results suggest that adaptive and neutral evolutionary processes have contributed to Oenothera diversification and rapid gene family evolution.

Satellite-based rapid sweeping screening of localized PM2.5 hotspots at fine-scale local neighborhood levels is highly desirable. This motivated us to develop a random forest–convolutional neural network–local contrast normalization (RF–CNN–LCN) pipeline that detects local PM2.5 hotspots at a 300 m resolution using satellite imagery and meteorological information. The RF–CNN joint model in the pipeline uses three meteorological variables and daily 3 m/pixel resolution PlanetScope satellite imagery to generate daily 300 m ground-level PM2.5 estimates. The downstream LCN processes the estimated PM2.5 maps to reveal local PM2.5 hotspots. The RF–CNN joint model achieved a low normalized root mean square error for PM2.5 of within ~31% and normalized mean absolute error of within ~19% on the holdout samples in both Delhi and Beijing. The RF–CNN–LCN pipeline reasonably predicts urban PM2.5 local hotspots and coolspots by capturing both the main intra-urban spatial trends in PM2.5 and the local variations in PM2.5 with urban landscape, with local hotspots relating to compact urban spatial structures and coolspots being open areas and green spaces. Based on 20 sampled representative neighborhoods in Delhi, our pipeline revealed an annual average 9.2 ± 4.0 μg m−3 difference in PM2.5 between the local hotspots and coolspots within the same community. In some cases, the differences were much larger; for example, at the Indian Gandhi International Airport, the increase was 20.3 μg m−3 from the coolest spot (the residential area immediately outside the airport) to the hottest spot (airport runway). This work provides a possible means of automatically identifying local PM2.5 hotspots at 300 m in heavily polluted megacities and highlights the potential existence of substantial health inequalities in long-term outdoor PM2.5 exposures even within the same local neighborhoods between local hotspots and coolspots.

Neuroprosthetic brain-computer interfaces function via an algorithm which decodes neural activity of the user into movements of an end effector, such as a cursor or robotic arm. In practice, the decoder is often learned by updating its parameters while the user performs a task. When the user's intention is not directly observable, recent methods have demonstrated value in training the decoder against a surrogate for the user's intended movement. Here we show that training a decoder in this way is a novel variant of an imitation learning problem, where an oracle or expert is employed for supervised training in lieu of direct observations, which are not available. Specifically, we describe how a generic imitation learning meta-algorithm, dataset aggregation (DAgger), can be adapted to train a generic brain-computer interface. By deriving existing learning algorithms for brain-computer interfaces in this framework, we provide a novel analysis of regret (an important metric of learning efficacy) for brain-computer interfaces. This analysis allows us to characterize the space of algorithmic variants and bounds on their regret rates. Existing approaches for decoder learning have been performed in the cursor control setting, but the available design principles for these decoders are such that it has been impossible to scale them to naturalistic settings. Leveraging our findings, we then offer an algorithm that combines imitation learning with optimal control, which should allow for training of arbitrary effectors for which optimal control can generate goal-oriented control. We demonstrate this novel and general BCI algorithm with simulated neuroprosthetic control of a 26 degree-of-freedom model of an arm, a sophisticated and realistic end effector.

This article addresses a collaborative writing experiment that explores spaces of diverse encounters that began at a research conference held in the Flamingo hotel in Las Vegas; spaces where knowings emerge in the (shared) moment, in-between (our)selves, prompted by different (research) questions and entanglements of matter and meaning. Through these multiple and emergent writing encounters we explore ways towards collaborative scholarly writing and accessible ways of working and knowing beyond the immediately known or sensed. In addition, this collaborative writing experiment serves to inspire and engage participants (qualitative researchers and ethnographers alike) to explore, share, and disseminate knowledge across contexts differently. We call for writing in qualitative research that senses, figures out, and “reveals” via moving and sensuous bodies, and emerging embodied encounters within particular spaces.

Foragers must often travel from a central place to exploit aggregations of prey. These patches can be identified behaviorally when a forager shifts from travel to area restricted search, identified by a decrease in speed and an increase in sinuosity of movement. Faster, more directed movement is associated with travel. Differentiating foraging behavior at patches from travel to patches is important for a variety of research questions and has now been made easier by the advent of small, GPS devices that can track forager movement with high resolution. In the summer and fall of 2012, movement data were collected from GPS devices placed on foraging trips originating in the artisanal fishing village of Desa Ikan (pseudonym), on the east coast of the Caribbean island nation of the Commonwealth Dominica. Moored FADs are human-made structures anchored to the ocean floor with fish attraction material on or near the surface designed to effectively create a resource patch. The ultimate goal of the research is to understand how property rights are emerging after the introduction of fish aggregating device (FAD) technology at the site in 1999. This paper reports on research to identify area-restricted search foraging behavior at FAD patches. For 22 foraging trips simultaneous behavioral observations were made to ground-truth the GPS movement data. Using a cumulative sum method, area restricted search was identified as negative deviations from the mean travel speed and the method was able to correctly identify FAD patches in every case.

During its formative months, I often faced the question whether this fourth IPY would include "kayaks" (my codeword for small, individual, and implicitly flexible and innovative research projects) or exclusively "icebreakers" (a codeword for large multi-investigator projects, by implication complex planned efforts, generally controlled by the established polar scientific order). The question of project scope arose in Arctic science meetings and penetrated, for a time, the public media. I maintained that innovation and creativity would occur throughout IPY, regardless of size of project or platform. As steward of the IPY International Office, I argued publicly for the widest inclusion of ideas and proposals, from groups large and small.

With the emergence of an entire generation of technologically infused learners comes a new challenge for educators to combat their desire for technologically enhanced isolation. The impact of social media on self-absorption and the ramifications of excessive screen time are already evident in our current generation of tech-native students. Meanwhile, the need for more empathetic employees who show compassion toward clients, exercise tolerance for co-workers and demonstrate cultural awareness in all aspects of their work is already being prioritised over discipline-specific skills and even STEM expertise by current business leaders. Similarly, the demand for more empathetic creatives is being catalysed by a design industry that now values human-centredness. Empathy brings the human component to user-experience and interface design and is now widely acknowledged as the first step in any design process. As a corollary, the development of empathy as a pedagogical deliverable will not only be necessary for the workplace of the future but for future generations to become fully-functioning members of society. This research explores how an inclusive art and design pedagogy can foster collaborative learner-centred approaches to develop empathy as a key component of emotional intelligence. Our methodology for triggering meaningful experiences and transformational growth harnesses collaborative play as a way to engage learners in stimulating, socially conscious teamwork. By becoming more intentional about designing the future of higher education in order to nurture more empathetic career creatives, we also lay the groundwork in preparation for educating the most technologically advanced, yet potentially socially isolated learners of any generation yet.