Now showing items 1-10 of 52
Deep Learning to Learn
(Georgia Institute of Technology, 2018-08-20)
Reinforcement learning and imitation learning have seen success in many domains, including autonomous helicopter flight, Atari, simulated locomotion, Go, robotic manipulation. However, sample complexity of these methods ...
Data-to-Decisions for Safe Autonomous Flight
(Georgia Institute of Technology, 2018-11-07)
Traditional sensor data can be augmented with new data sources such as roadmaps and geographical information system (GIS) Lidar/video to offer emerging unmanned aircraft systems (UAS) and urban air mobility (UAM) a new ...
TF-Slim: A Lightweight Library for Defining, Training and Evaluating Complex Models in TensorFlow
(Georgia Institute of Technology, 2017-09-07)
TF-Slim is a TensorFlow-based library with various components. These include modules for easily defining neural network models with few lines of code, routines for training and evaluating such models in a highly distributed ...
Reaching Beyond Human Accuracy With AI Datacenters
Deep learning has enabled rapid progress in diverse problems in vision, speech, healthcare, and beyond. This progress has been driven by breakthroughs in algorithms that can harness massive datasets and powerful compute ...
Automated Perception in the Real World: The Problem of Scarce Data
Machine perception is a key step toward artificial intelligence in domains such as self-driving cars, industrial automation, and robotics. Much progress has been made in the past decade, driven by machine learning, ...
Structured Prediction - Beyond Support Vector Machine and Cross Entropy
Many classification tasks in machine learning lie beyond the classical binary and multi-class classification settings. In those tasks, the output elements are structured objects made of interdependent parts, such as sequences ...
Do GANs Actually Learn the Distribution?
(Georgia Institute of Technology, 2018-02-22)
Generative Adversarial Nets (GANs) is a framework for training deep generative models, due to Goodfellow et al'13. It involves a competition between a generator net that tries to produce realistic images, and a discriminator ...
Learning Locomotion: From Simulation to Real World
Deep Reinforcement Learning (DRL) holds the promise of designing complex robotic controllers automatically. In this talk, I will discuss two different approaches to apply deep reinforcement learning to learn locomotion ...
The Geometry of Community Detection via the MMSE Matrix
The information-theoretic limits of community detection have been studied extensively for network models with high levels of symmetry or homogeneity. In this talk, Reeves will present a new approach that applies to a broader ...
Deep Networks for Pixel Level Inference with Applications to Medical Imaging
(Georgia Institute of Technology, 2017-09-26)