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.

Transmigration in Indonesia: Lessons from its Environmental and Social Impacts

Abstract: This article identifies the lessons learned from transmigration programs in Indonesia during 1976-89 and describes briefly the history and types of transmigration in Indonesia the World Bank project the demographic and agricultural benefits and the environmental and social impacts. During transmigration millions of people from overcrowded islands of Java Madura Bali and Lombok were resettled in the outer islands of Sumatra Kalimantan Sulawesi and Irian Jaya. The World Bank which funded the program has been criticized for its irresponsibility. An evaluation of the relative benefits of resettlement schemes is dependent upon answering several questions. One question is to what extent development initiatives "going wrong" should be accepted and given support to lesson the damage. Another question is to what extent should financing agencies be responsible for damage that is unlike limited impacts of more discrete projects. About 17% of transmigration projects are corrupt and choice of sites is controversial. Environmental impact statements are required but are not publicly available or debated. Impact assessments stipulate inclusion of local people in the process whereas in practice locals are included as data. Sometimes impacts are not ready before industry is installed. World Bank review processes result in significant deletions between draft and final Appraisal Reports. Governments maintain secrecy. The effect of transmigration is the diluting of native cultures the excuse-making because "it is going to happen anyway" inadequate assessments environmental degradation and continuance of schemes under other names.

Deep Active Learning for Named Entity Recognition.

Abstract: Deep learning has yielded state-of-the-art performance on many natural language processing tasks including named entity recognition (NER). However, this typically requires large amounts of labeled data. In this work, we demonstrate that the amount of labeled training data can be drastically reduced when deep learning is combined with active learning. While active learning is sample-efficient, it can be computationally expensive since it requires iterative retraining. To speed this up, we introduce a lightweight architecture for NER, viz., the CNN-CNN-LSTM model consisting of convolutional character and word encoders and a long short term memory (LSTM) tag decoder. The model achieves nearly state-of-the-art performance on standard datasets for the task while being computationally much more efficient than best performing models. We carry out incremental active learning, during the training process, and are able to nearly match state-of-the-art performance with just 25\% of the original training data.

Channel Attention Is All You Need for Video Frame Interpolation

Abstract: Prevailing video frame interpolation techniques rely heavily on optical flow estimation and require additional model complexity and computational cost; it is also susceptible to error propagation in challenging scenarios with large motion and heavy occlusion. To alleviate the limitation, we propose a simple but effective deep neural network for video frame interpolation, which is end-to-end trainable and is free from a motion estimation network component. Our algorithm employs a special feature reshaping operation, referred to as PixelShuffle, with a channel attention, which replaces the optical flow computation module. The main idea behind the design is to distribute the information in a feature map into multiple channels and extract motion information by attending the channels for pixel-level frame synthesis. The model given by this principle turns out to be effective in the presence of challenging motion and occlusion. We construct a comprehensive evaluation benchmark and demonstrate that the proposed approach achieves outstanding performance compared to the existing models with a component for optical flow computation.

Providing virtual networking functionality for managed computer networks

Abstract: Techniques are described for providing virtual networking functionality for managed computer networks. In some situations, a user may configure or otherwise specify one or more virtual local area networks (“VLANs”) for a managed computer network being provided for the user, such as with each VLAN including multiple computing nodes of the managed computer network. Networking functionality corresponding to the specified VLAN(s) may then be provided in various manners, such as if the managed computer network itself is a distinct virtual computer network overlaid on one or more other computer networks, and communications between computing nodes of the managed virtual computer network are handled in accordance with the specified VLAN(s) of the managed virtual computer network by emulating functionality that would be provided by networking devices of the managed virtual computer network if they were physically present and configured to support the specified VLAN(s).