arXiv 2020

MedDialog: A Large-scale Medical Dialogue Dataset

arXiv 2020 UCSD-AI4H/Medical-Dialogue-System

Medical dialogue systems are promising in assisting in telemedicine to increase access to healthcare services, improve the quality of patient care, and reduce medical costs.

MedDialog: A Large-scale Medical Dialogue Dataset

arXiv 2020 UCSD-AI4H/Medical-Dialogue-System

Medical dialogue systems are promising in assisting in telemedicine to increase access to healthcare services, improve the quality of patient care, and reduce medical costs.

Simulation-based Lidar Super-resolution for Ground Vehicles

arXiv 2020 RobustFieldAutonomyLab/lidar_super_resolution

We propose a methodology for lidar super-resolution with ground vehicles driving on roadways, which relies completely on a driving simulator to enhance, via deep learning, the apparent resolution of a physical lidar.

IMAGE SUPER-RESOLUTION

PolyLaneNet: Lane Estimation via Deep Polynomial Regression

arXiv 2020 lucastabelini/PolyLaneNet

One of the main factors that contributed to the large advances in autonomous driving is the advent of deep learning.

AUTONOMOUS DRIVING LANE DETECTION

PolyLaneNet: Lane Estimation via Deep Polynomial Regression

arXiv 2020 lucastabelini/PolyLaneNet

One of the main factors that contributed to the large advances in autonomous driving is the advent of deep learning.

AUTONOMOUS DRIVING LANE DETECTION

Understanding Integrated Gradients with SmoothTaylor for Deep Neural Network Attribution

arXiv 2020 garygsw/smooth-taylor

Integrated gradients as an attribution method for deep neural network models offers simple implementability.

IMAGE CLASSIFICATION OBJECT RECOGNITION

Understanding Integrated Gradients with SmoothTaylor for Deep Neural Network Attribution

arXiv 2020 garygsw/smooth-taylor

Integrated gradients as an attribution method for deep neural network models offers simple implementability.

IMAGE CLASSIFICATION OBJECT RECOGNITION

Supervised Domain Adaptation: Were we doing Graph Embedding all along?

arXiv 2020 lukashedegaard/dage

In this paper, we show that existing Domain Adaptation methods can be formulated as Graph Embedding methods in which the domain labels of samples coming from the source and target domains are incorporated into the structure of the intrinsic and penalty graphs used for the embedding.

DOMAIN ADAPTATION GRAPH EMBEDDING

Quantum Approximation for Multi-Scale Scheduling

arXiv 2020 lytzV/QAOA

Then, for the given MWIS, the proposed QAOS designs the Hamiltonian of the problem.

QUANTUM APPROXIMATE OPTIMIZATION