Transfer Learning for Human & AI

A large class of ergodic control algorithms relies on spectral analysis, which suffers from the curse of dimensionality . The proposed solution is efficient both computationally and storage-wise, hencemaking it suitable for its online implementation in robotic systems . The approach is applied to a peg-in-hole insertion task using a 7-axis Franka EmikaPanda robot, where the task can be achieved without the use of force/torque sensors. The approach allows the task to be done without the need for force or torque sensors. It also allows the robot to be able to insert a peg in a hole without using force or torque sensors in order to find the hole in the hole. The proposal is based on tensor traindecomposition, a recent approach from multilinear algebra for efficient computation of multidimensional arrays. The

Author(s) : Suhan Shetty, João Silvério, Sylvain Calinon

Links : PDF - Abstract

Code :

https://github.com/nhynes/abc

Keywords : hole - approach - force - torque - sensors -

• AI for Everyone by Andrew Ng
• School of Data Science
• Machine Learning A-Z
• Kaggle Competitions
• Data Science A-Z
• Khan Academy Multivariable Calculus
• Deep Learning Nanodegree
• Introduction to R