Explore the vast range of titles available.

Surrogate machine-learning models are transforming computational materials science by predicting properties of materials with the accuracy of ab initio methods at a fraction of the computational cost. We demonstrate surrogate models that simultaneously interpolate energies of different materials on a dataset of 10 binary alloys (AgCu, AlFe, AlMg, AlNi, AlTi, CoNi, CuFe, CuNi, FeV, and NbNi) with 10 different species and all possible fcc, bcc, and hcp structures up to eight atoms in the unit cell, 15,950 structures in total. We find that the deviation of prediction errors when increasing the number of simultaneously modeled alloys is <1 meV/atom. Several state-of-the-art materials representations and learning algorithms were found to qualitatively agree on the prediction errors of formation enthalpy with relative errors of <2.5% for all systems.

Various approaches have used neural networks as probabilistic models for the design of protein sequences. These \"inverse folding\" models employ different objective functions, which come with trade-offs that have not been assessed in detail before. This study introduces probabilistic definitions of protein stability and conformational specificity and demonstrates the relationship between these chemical properties and the p ( structure | seq ) Boltzmann probability objective. This links the Boltzmann probability objective function to experimentally verifiable outcomes. We propose a novel sequence decoding algorithm, referred to as “BayesDesign”, that leverages Bayes’ Rule to maximize the p ( structure | seq ) objective instead of the  p ( seq | structure ) objective common in inverse folding models. The efficacy of BayesDesign is evaluated in the context of two protein model systems, the NanoLuc enzyme and the WW structural motif. Both BayesDesign and the baseline ProteinMPNN algorithm increase the thermostability of NanoLuc and increase the conformational specificity of WW. The possible sources of error in the model are analyzed.

Human teams are able to easily perform collaborative manipulation tasks. However, simultaneously manipulating a large extended object for a robot and human is a difficult task due to the inherent ambiguity in the desired motion. Our approach in this paper is to leverage data from human-human dyad experiments to determine motion intent for a physical human-robot co-manipulation task. We do this by showing that the human-human dyad data exhibits distinct torque triggers for a lateral movement. As an alternative intent estimation method, we also develop a deep neural network based on motion data from human-human trials to predict future trajectories based on past object motion. We then show how force and motion data can be used to determine robot control in a human-robot dyad. Finally, we compare human-human dyad performance to the performance of two controllers that we developed for human-robot co-manipulation. We evaluate these controllers in three-degree-of-freedom planar motion where determining if the task involves rotation or translation is ambiguous.

The 6.5 Ha Nonsuch island nature reserve in Castle harbour, Bermuda, under restoration as a “Living museum” of Bermuda’s pre-colonial terrestrial flora and fauna since 1962, was unexpectedly compromised by the colonization of a manmade freshwater pond by the invasive Cane toad Bufo marinus, which managed to cross a 175 m seawater barrier isolating the island. The toad was considered to pose a threat to the endemic Skink, for which Nonsuch is a key sanctuary, and to the burrow nesting Bermuda petrel or Cahow, which was about to be translocated to that larger island in order to establish a colony safe from the global warming threat of sealevel rise. A non-lethal method of removing the toads was devised through installation of a toad barrier around the freshwater pond, constructed of robust high density polyethylene (HDPE). Toads outside the barrier were no longer able to breed and were captured and removed by night searches over a five year period as they piled up against it. Removal of the breeding population inside the barrier was only possible by capturing all individuals resulting from each spawning before they could reach breeding age. This also took 5 years despite the small size (0.06 Ha) of the pond. A total of 1,244 toads were removed to Bermuda’s mainland. As breeding age is attained in 2 years the collection total for the first 2 years provided an approximation of the population density on the island before removal began (~80 per Ha). Although the immigration rate to Nonsuch is estimated to be less than one per year, this still requires maintaining the barrier as a permanent fixture. As future immigrants can be fairly quickly intercepted, the toad is now essentially extaxic on Nonsuch. This is the first published report of a successful eradication of the species from an island with breeding habitat.

Breeding Common Tems (Sterna hirundo) have been studied extensively in North America and Europe, but little is known of the small, isolated populations south of the main continental ranges. We conducted a 69 year study of Common Terns in the Bermuda archipelago (in the western North Atlantic Ocean at 32[degrees]N) to document their phenology, distribution, numbers, and breeding biology. Common Terns arrive at Bennuda in March and start to nest about 10 April, 2-3 weeks earlier than any other population of the species studied to date. Unlike the colonially nesting mainland populations, Bermudian Common Terns nest singly or in very small groups on many small islets and artificial nest sites (n = 50 sites) scattered throughout the archipelago. The population included 15-36 pairs in the 1970s and 1980s, but declined by about 50% following each of 3 major hurricanes in 1987, 2003, and 2010, and is now critically endangered. Since the 2003 hurricane, the population has included many (up to 7) female-female pairs laying infertile eggs. In contrast, the mean clutch size of male-female pairs was 3.01, with mean productivity >2.2 fledglings/pair over the 69 year period, the highest values reported for the species anywhere in its range; in at least 10 cases, pairs raised 4 chicks to fledging. The Bermudian population of Common Tems is intensively managed but remains critically endangered, and its recent shift to nesting on small rafts and buoys about 3 m in diameter poses new challenges for management. Received 6 February 2019. Accepted 10 August 2020.

The movements and behavior of many taxa of seabirds during the non-breeding season remain poorly known. For example, although studies conducted in the Pacific and Indian oceans suggest that White-tailed Tropicbirds (Phaethon lepturus) seldom fly more than a few thousand kilometers from nest colonies after breeding, little is known about the post-breeding movements and behavior of a subspecies of White-tailed Tropicbirds (P. l. catesbyi) that breeds on islands in the North Atlantic Ocean. Our objective therefore, was to use light-based geolocators to identify the ranges and pelagic activities of White-tailed Tropicbirds from Bermuda during the non-breeding periods in 2014—2015 (TV—25) and 2015—2016 (N=16). Locations were estimated based on changes in light intensity across time, and pelagic activities were determined based on whether geolocators attached to leg bands were wet (i.e., birds resting on the water's surface) or dry (i.e., birds in flight). In 2014, birds spent late summer (July-September) near Bermuda and the British Virgin Islands; by mid-September, most (N = 17; 68%) birds took a direct easterly route to the Sargasso Sea. In 2015, most post-breeders (N = 15; 94%) flew east from Bermuda and to the Sargasso before the end of late summer. For both years combined, fall and winter (October-February) ranges extended as far west as North Carolina and as far east as the mid-Atlantic Ridge. In both years, all birds were located between Bermuda and the British Virgin Islands during the spring (April-May). All birds then flew north to Bermuda in both years, with variations in timing, during April and May. We also found extensive overlap in the ranges of males and females during the non-breeding season in both years. During the non-breeding season, Whitetailed Tropicbirds spent 5% of night periods and 41% of day periods in flight in 2014; in 2015, birds spent 8% and 42% of night and day periods, respectively, in flight. Tropicbirds spent more time flying during the day because they hunt by day, detecting prey on the wing by sight. Overall, our results suggest that Whitetailed Tropicbirds that breed in Bermuda are diurnal, nomadic wanderers that range over an extensive area of the Atlantic Ocean during the non-breeding season. Los movimientos y el comportamiento de muchos taxones de aves marinas durante la temporada no reproductiva permanecen todavía pobremente conocidos. Por ejemplo, aun cuando estudios realizados en los océanos Pacífico e Indico sugieren que el Rabijunco común (Phaethon lepturus) rara vez vuela más de unos pocos miles de kilómetros de la colonia de nidada luego de reproducir, poco se sabe sobre los movimientos y el comportamiento post reproductivos de una subespecie de Rabijunco común (Paethon l. castebyi) que reproduce en islas del Océano Atlántico Norte. Nuestro objetivo, entonces, fue usar geolocalizadores basados en luz para identificar los rangos de distribución y las actividades pelagicas del Rabijunco común de Bermuda durante los períodos no reproductivos en 2014-2015 (N = 25) y en 2015-2016 (N = 16). Se estimaron las ubicaciones a partir de los cambios en la intensidad de luz a lo largo del tiempo, y las actividades pelágicas fueron determinadas a partir de si los geolocalizadores adheridos a las bandas de las patas estaban mojados (i.e., aves descansando en la superficie del agua) secos (i.e. aves volando). En 2014, las aves pasaron el verano tardío (julio-septiembre) cerca de Bermuda y las Islas Vírgenes Británicas; para mediados de septiembre, la mayoría (N = 17, 68%) de las aves tomaron una ruta directa y hacia el este al Mar de los Sargazos. En 2015, la mayoría de los individuos post reproductivos (N = 15; 94%) volaron hacia el este desde Bermuda y hacia el Mar de los Sargazos antes del final del verano. Para ambos años combinados, los rangos en otoño e invierno (octubre-febrero) se extendieron hacia el oeste tan lejos como Carolina del Norte y hacia el este tan lejos como la Dorsal Mesoatlántica. En ambos años, todas las aves fueron ubicadas entre Bermuda y las Islas Vírgenes Británicas durante la primavera (abril-mayo). Todas las aves luego volaron norte a Bermuda en ambos años, con variaciones en el momento, durante abril y mayo. También encontramos extensa superposición en los rangos de los machos y las hembras durante la temporada no reproductiva en ambos años. Durante la temporada no reproductiva, los Rabijuncos comunes pasaron el 5% de los períodos nocturnos y el 41% de los períodos diurnos en vuelo en 2014; en 2015, las aves pasaron el 8% y el 42% de los períodos nocturnos y diurnos, respectivamente, en vuelo. Los Rabijuncos pasaron más tiempo volando durante el dia debido a que cazan de dia, detectando las presas en vuelo medio de la vista. En general, nuestros resultados sugieren que los Rabijuncos comunes que crían en Bermuda son diurnos, nómades errantes que se distribuyen en una extensa área del Océano Atlàntico durante la temporada no reproductiva.

Breeding Common Terns (Sterna hirundo) have been studied extensively in North America and Europe, but little is known of the small, isolated populations south of the main continental ranges. We conducted a 69 year study of Common Terns in the Bermuda archipelago (in the western North Atlantic Ocean at ∼32°N) to document their phenology, distribution, numbers, and breeding biology. Common Terns arrive at Bermuda in March and start to nest about 10 April, 2–3 weeks earlier than any other population of the species studied to date. Unlike the colonially nesting mainland populations, Bermudian Common Terns nest singly or in very small groups on many small islets and artificial nest sites (n = 50 sites) scattered throughout the archipelago. The population included 15–36 pairs in the 1970s and 1980s, but declined by about 50% following each of 3 major hurricanes in 1987, 2003, and 2010, and is now critically endangered. Since the 2003 hurricane, the population has included many (up to 7) female–female pairs laying infertile eggs. In contrast, the mean clutch size of male–female pairs was 3.01, with mean productivity >2.2 fledglings/pair over the 69 year period, the highest values reported for the species anywhere in its range; in at least 10 cases, pairs raised 4 chicks to fledging. The Bermudian population of Common Terns is intensively managed but remains critically endangered, and its recent shift to nesting on small rafts and buoys about 3 m in diameter poses new challenges for management.

In Bermuda, only three species of breeding seabirds survived four centuries of human activity, dating back to the settlement of the archipelago in 1609. This paper reports two consecutive years of nesting data on the Roseate Tern (Sterna dougallii), a historic breeder in Bermuda last recorded breeding in 1849. A pair was discovered incubating an egg on 7 July 2018 on Pearl Island, and a pair, plus one additional adult, returned to this islet to breed in 2019. Roseate Terns and nests were intensively monitored using offshore observations and video recordings. In 2018, the single egg hatched on 24 July and the chick fledged on 18 August (fledging period = 25 d). In 2019, we found a two-egg clutch on 4 June, which hatched on 15 and 17 June. The two chicks were flying by 13 July (fledging periods = 26-28 d). The breeding Roseate Terns in Bermuda are probably derived from the Caribbean/Bahamas population. The two-year nest-site fidelity of Roseate Terns, following absence for at least 140 years, suggests that this species is potentially reestablishing itself as a breeder in Bermuda. Future studies should determine the genetic origin of the birds, determine if the observed pairs are the same individuals using leg bands, record population growth, and compare the nesting biology of Bermudian Roseate Terns to that of the Bermudian Common Terns (Sterna hirundo).

We propose and explore the possibility that language models can be studied as effective proxies for specific human subpopulations in social science research. Practical and research applications of artificial intelligence tools have sometimes been limited by problematic biases (such as racism or sexism), which are often treated as uniform properties of the models. We show that the “algorithmic bias” within one such tool—the GPT-3 language model—is instead both fine-grained and demographically correlated, meaning that proper conditioning will cause it to accurately emulate response distributions from a wide variety of human subgroups. We term this property algorithmic fidelity and explore its extent in GPT-3. We create “silicon samples” by conditioning the model on thousands of sociodemographic backstories from real human participants in multiple large surveys conducted in the United States. We then compare the silicon and human samples to demonstrate that the information contained in GPT-3 goes far beyond surface similarity. It is nuanced, multifaceted, and reflects the complex interplay between ideas, attitudes, and sociocultural context that characterize human attitudes. We suggest that language models with sufficient algorithmic fidelity thus constitute a novel and powerful tool to advance understanding of humans and society across a variety of disciplines.

Breeding cavity-nesting seabirds on remote, oceanic islands are threatened largely by destruction of nest cavities and introduced predators. In response, artificial cavities are often implemented as a conservation strategy. Effective artificial cavity programs require recognition of specific nest-site characteristics, which can be major determinants of survival and persistence of cavity-nesting seabirds. In Bermuda, we monitored 158 natural and 178 artificial cavities of the White-tailed Tropicbird (Phaethon lepturus catesbyi) from 2013–2015 to determine if specific nest-site characteristics could explain cavity selection, nest survival, and predation by introduced rats (Rattus spp.) and the American Crow (Corvus brachyrhynchos). We found that tropicbirds preferred natural cavities lined with sand, and which had smaller entrances and were located on steeper cliffs. Occupancy of artificial cavities on the mainland varied across years and increased with cavity depth, whereas neither variable had any effect on artificial cavity occupancy on satellite islands. Clutch survival declined in cavities with rubble in nesting sand, and those with shallower cavity depths and larger entrance heights. Nestling survival varied by year, declined as the nesting season progressed, and increased with nestling age. Rat predation was significantly higher on mainland sites compared with satellite islands. Crow predation occurred only on offshore island sites, where natural cavities and shallower cavities with larger entrance heights were vulnerable. Rat and crow predation rates varied across years. Our results suggest multiple nest-site characteristics and temporal factors are important predictors of productivity in White-tailed Tropicbirds. Conservationists should simultaneously consider nest-cavity siting and dimensions, while they conduct biosecurity and predator control to improve the nest success of cavity-nesting seabirds.