Amazon.com

About: Amazon.com is a company organization based out in Seattle, Washington, United States. It is known for research contribution in the topics: Service (business) & Service provider. The organization has 13363 authors who have published 17317 publications receiving 266589 citations.

Dynamically generating haptic effects from audio data

Abstract: Haptic effects are dynamically generated for content presentation on a device through analysis of the content. During content playback, audio data for the content may be analyzed to determine low frequency audio data. The low frequency audio data is mapped from a low frequency range to a haptic control frequency range of one or more haptic actuators included in the device. This mapping may be used to generate a control signal to drive the one or more haptic actuators. The haptic effects and the content may be synchronized to one another during the presentation of the content on the device. The haptic actuator control signal may be amplified proportionally to the amplitude of the low frequency audio data.

Learning Generative ConvNets via Multi-grid Modeling and Sampling

Abstract: This paper proposes a multi-grid method for learning energy-based generative ConvNet models of images. For each grid, we learn an energy-based probabilistic model where the energy function is defined by a bottom-up convolutional neural network (ConvNet or CNN). Learning such a model requires generating synthesized examples from the model. Within each iteration of our learning algorithm, for each observed training image, we generate synthesized images at multiple grids by initializing the finite-step MCMC sampling from a minimal 1 A— 1 version of the training image. The synthesized image at each subsequent grid is obtained by a finite-step MCMC initialized from the synthesized image generated at the previous coarser grid. After obtaining the synthesized examples, the parameters of the models at multiple grids are updated separately and simultaneously based on the differences between synthesized and observed examples. We show that this multi-grid method can learn realistic energy-based generative ConvNet models, and it outperforms the original contrastive divergence (CD) and persistent CD.

Achieving zero deforestation in the Brazilian Amazon: What is missing?

Abstract: Amazon deforestation causes severe climatic and ecological disruptions, with negative consequences for the livelihood of forest-dependent peoples. To avoid further disruptions, Brazil will need to take bold steps to eliminate both illegal and legal Amazon deforestation over the short term. Amazon deforestation declined by 70% between 2005 and 2014 due to drops in commodity prices and interventions by federal and state governments, such as law enforcement campaigns and credit restrictions for landowners who deforest illegally. Despite these impressive achievements, Brazil still deforests 5,000 km2 of Amazonian forests each year. How then will Brazil eliminate Amazon deforestation altogether if the country is only committed to cut illegal deforestation by 2030—as stated in its Intended Nationally Determined Contributions (iNDC) to the 2015 climate change treaty meeting in Paris? Here we provide an analysis of the major socio-economic-political threats that could constrain Brazil from achieving its current goals. We then propose six fundamental strategies to help Brazil achieve a more ambitious goal to eliminate all major legal and illegal Amazon deforestation. These strategies involve bringing social and environmental safeguards to the infrastructure plans in the region, consolidating and expanding positive incentives for the production of sustainable commodities, establishing a new policy to guarantee the social and environmental sustainability of rural settlements, fully implementing the national legislation protecting forests (the Forest Code), protecting the land rights of indigenous people and traditional communities, and expanding the existing network of protected areas, allocating the 80 million hectares of not designated public forests as protected areas or areas for sustainable use of timber and non-timber forest products. The implementation of these strategies however depends on the formulation of a new development paradigm that promotes economic growth, social justice and productive agriculture, while protecting the fundamentally important ecological services of tropical forests.

Trans-Amazonian natal homing in giant catfish

Abstract: Knowledge of fish migration is a prerequisite to sustainable fisheries management and preservation, especially in large international river basins. In particular, understanding whether a migratory lifestyle is compulsory or facultative, and whether adults home to their natal geographic area is paramount to fully appraise disruptions of longitudinal connectivity resulting from damming. In the Amazon, the large migratory catfishes of the Brachyplatystoma genus are apex predators of considerable interest for fisheries. They are believed to use the entire length of the basin to perform their life cycle, with hypothesized homing behaviours. Here, we tested these hypotheses, using the emblematic B. rousseauxii as a model species. We sampled adults close to major breeding areas in the Amazon basin (upper Madeira and upper Amazonas) and assessed their lifetime movements by measuring variations in 87Sr/86Sr along transverse sections of their otoliths (ear stones) using laser ablation multicollector mass spectrometry (LA-MC-ICPMS). We demonstrate that larvae migrate downstream from the Andean piedmont to the lower Amazon, where they grow over a protracted period before migrating upstream as adults. Contrary to prevailing inferences, not all fish spend their nursery stages in the Amazon estuary. By contrast, the passage in the lower or central Amazon seems an obligate part of the life cycle. We further evidence that most adults home to their natal geographic area within the Madeira sub-basin. Such long-distance natal homing is exceptional in purely freshwater fishes. Synthesis and applications. By using otolith microchemistry, we were able to demonstrate a seemingly compulsory basin-wide migratory life cycle of large Amazonian catfishes. This makes them the organisms performing the longest migrations (>8000 km) in fresh waters. This exceptional life history is already jeopardized by two dams recently built in the Madeira River, which block a major migration route and access to a substantial part of their spawning grounds. Major impacts can be anticipated from the current and forthcoming hydroelectric development in the Amazon basin, not only on the populations and fisheries of this apex predator, but also on Amazonian food webs through trophic cascades.

Threading as a service

Abstract: A service manages a plurality of virtual machine instances for low latency execution of user codes. The plurality of virtual machine instances can be configured based on a predetermined set of configurations. One or more containers may be created within the virtual machine instances. In response to a request to execute user code, the service identifies a pre-configured virtual machine instance suitable for executing the user code. The service can allocate the identified virtual machine instance to the user, create a new container within an instance already allocated to the user, or re-use a container already created for execution of the user code. When the user code has not been activated for a time-out period, the service can invalidate allocation of the virtual machine instance destroy the container. The time from receiving the request to beginning code execution is less than a predetermined duration, for example, 100 ms.