The Future of Robotics

4min Snip

00:00

Play full episode

Summary

Transcript

Episode notes

There wasn't any deep learning in this paper at all. And so this is not exactly how to, we don't need deep learning to really move robotics forward still. There's still so much work to be done. So the main focus was on this sort of classical optimization technique. It actually got me curious to look at the other papers that were nominated and we won't go too much into them,. But here's just a few.

Our 99th episode with a summary and discussion of last week's big AI news!

Sponsor: This episode is sponsored by Zencastr, our go-to tool for recording the podcast. It is super easy to use, and there is nothing to download. Go to http://zen.ai/lastweekinai and get 30% off your first three months with a PRO account!

Outline:

(00:00) Intro
(1:50) Royal Mail is Doing the Right Thing with Drone Delivery
(6:10) Google uses AI to digitise maps in the utilities industry

(8:57) AI on the Ball: Startup Shoots Computer Vision to the Soccer Pitch

(9:26) Google Adds Machine Learning-Powered Studio Lighting To Video Calls

(11:33) Dyson launches a hiring spree for hundreds of engineers to build robots capable of doing household chores

(12:20) Pony.ai loses permit to test autonomous vehicles with driver in California

(12:52) Researchers develop algorithm to divvy up tasks for human-robot teams

(19:42) Microsoft AI Researchers Introduces (De)ToxiGen: A Large-Scale Machine-Generated Dataset for Adversarial and Implicit Hate Speech Detection

(24:32) Researchers Use Machine Learning to Identify Thousands of New Marine RNA Viruses in Study of Interest to Microbiologists and Clinical Laboratory Scientists

(25:17) Tiny robotic crab is smallest-ever remote-controlled walking robot

(26:12) TartanDrive dataset likely largest for off-road environments

(26:48) AI Framework Cuts Time, Effort Needed to Predict Critical Care Outcomes