Despite the well-developed cut-edge representation learning for language, most language representation models usually focus on specific level of linguistic unit, which cause great inconvenience when being confronted w...
详细信息
Out-of-distribution (OOD) generalization in the graph domain is challenging due to complex distribution shifts and a lack of environmental contexts. Recent methods attempt to enhance graph OOD generalization by genera...
详细信息
Recognizing novel sub-categories with scarce samples is an essential and challenging research topic in computer vision. Existing literature addresses this challenge by employing local-based representation approaches, ...
详细信息
Evidence suggests that the neural system associated with face processing is a distributed cortical network containing both bottom-up and top-down mechanisms. While bottom-up face processing has been the focus of many ...
详细信息
The advancements in sensor technology have made it possible to design wearable devices specifically designed for animals. These wearable devices can be used for locating individual animals, monitor their status, and t...
详细信息
ISBN:
(纸本)9781728138862
The advancements in sensor technology have made it possible to design wearable devices specifically designed for animals. These wearable devices can be used for locating individual animals, monitor their status, and track their trajectories in the wild. Some animal groups (such as chimpanzees) exhibit complex group behavior and these group dynamics play an important role in the physical and mental health of the animals. Scientists have traditionally been monitoring group dynamics manually in the wild. This requires extensive field trips, costing a lot of time and money. This calls for using the recent developments in technology, such as smart wearable devices for this purpose. However, lack of infrastructure support (limited connectivity, limited power, etc.) in the wilderness makes this a tedious task. In this work-in-progress paper, we present our technological approach and how we address the issues of wilderness to study animal behavior. We demonstrate how we build a network of lightweight wearable devices, and how the digital output of these devices can be used to analyze animal relationship. We show an initial, exploratory experiment, outlining the capabilities of the devices and technologies used in terms of communication efficiency, and the potential of the devices that can be used in the wilderness. Our initial results show that up to 90% of the proximity-based interactions can be captured.
In this paper, we propose a deep kernel embedded clustering network, namely DKEC, which learns data partitions with kernelized semantic embeddings of data samples via a self-supervised deep neural network. A kernelize...
详细信息
This work discusses a trajectory tracking task accomplished by a group of three Bebop 2 quadrotors working cooperatively to transport a load. Such unmanned aerial vehicles (UAVs) are considered as a triangular formati...
详细信息
ISBN:
(数字)9781728142784
ISBN:
(纸本)9781728142791
This work discusses a trajectory tracking task accomplished by a group of three Bebop 2 quadrotors working cooperatively to transport a load. Such unmanned aerial vehicles (UAVs) are considered as a triangular formation, and the load, a triangle-shaped structure, is attached to them through massless flexible cables. The reason to use three quadrotors is the gain in terms of the payload capability and the stability of the load, since it is attached to three vehicles, staying vertically aligned with the center of mass of the triangle correspondent to the formation. A formation controller, actually a kinematic controller applied to a virtual massless robot correspondent to the center of mass of the triangle, is adopted, which generates reference velocities for the three UAVs, whose movement generates the movement and reshaping of the triangular formation. Then, individual dynamic compensators are adopted for each UAV, whose input is the reference velocity delivered by the formation controller. Finally, experimental results are discussed, which validate the proposal.
Exploiting common language as an auxiliary for better translation has a long tradition in machine translation, which lets supervised learning based machine translation enjoy the enhancement delivered by the well-used ...
详细信息
Deep Learning models like Convolutional Neural Networks (CNN) are powerful image classifiers, but what factors determine whether they attend to similar image areas as humans do? While previous studies have focused on ...
详细信息
暂无评论