BTW, @Tesla employes a data labelling team of over 1,000 people in-house. This is not part-time role. They are professionals working on a custom-built data labelling infrastructure, side by side with engineers.

Next time you wonder about the scalability of supervised learning.

I highly recommend you watching it. I’m just 30min in, and it’s highly fascinating to see how ML has been deployed to solve the various problems of self-driving. It’s also fascinating to see how the approach is evolving and what’s their expectations going forward. https://t.co/SZIUhQ6k3q

Reviewing the @Tesla #AI Day and a lot of funny questions come to mind.

For example: what happens in a sales parking lot with 200 identical white cars?

Or: how does the car behave when, while driving, 2 vehicles have an accident in the opposite lane?

https://t.co/6JhQdNFM2Z

I’m also running the same poll on LinkedIn (but for a longer timeframe).

I’ll update on the results and compare. https://t.co/mx1PzxmrVG

RT @giano: How many *paid* newsletters do you subscribe to?

Every week I ask the following question: “What is the book that changed your life?” and, every week, I publish the answers I get.
This week:

– On the Social Contract
– The Federalist Papers
– Violence

Full list: https://t.co/pgzzIv3ajy https://t.co/Jzy8EC3Nie

“If your customer service strategy consists of mollifying angry customers, you’ll always be behind.”

https://t.co/QMVzwLITWg

How many *paid* newsletters do you subscribe to?

Interesting question and fascinating answers that I encourage you to read. Some requests are highly impractical. Others, liked by many respondents, could become a reality with the right collaboration in place. https://t.co/UDRFwrQBhO

@virtualmartijn In every decade there have been 70yo with not enough money for retirement. The difference, and the point, is how radically different the remediation plan has become in my example.

And no, this is absolutely not a hobby. I had a long conversation and he explained me in details.