One-Shot Imitation from Observing Humans via Domain-Adaptive Meta-Learning
Offered By: University of Central Florida via YouTube
Course Description
Overview
Explore cutting-edge approaches to imitation learning in this 30-minute lecture from the University of Central Florida. Delve into the concept of one-shot imitation from human observation through domain-adaptive meta-learning. Examine the problem definition, training objectives, and key algorithms, including the policy network and temporal convolution network. Investigate the application of spatial softmax and its role in task execution. Analyze experiments involving large domain shifts and their results. Gain insights into the latest developments in this field and understand their potential implications for future AI applications.
Syllabus
Intro
Outline
Approaches for Imitation Learning
Dataset
Problem Definition
Training Objective
Algorithm 1
The Policy Network
Temporal Convolution Network
Spatial Softmax
Tasks
Some subset of objects
Experiments
Large Domain Shift
Results
Conclusions
Taught by
UCF CRCV
Tags
Related Courses
Introduction to Deep LearningMassachusetts Institute of Technology via YouTube Taming Dataset Bias via Domain Adaptation
Alexander Amini via YouTube Making Our Models Robust to Changing Visual Environments
Andreas Geiger via YouTube Learning Compact Representation with Less Labeled Data from Sensors
tinyML via YouTube Geo-localization Framework for Real-world Scenarios - Defense Presentation
University of Central Florida via YouTube