Copy
Want to be a campus rep? Reply to our email

July 22, 2020

 

 👋 <<First Name>>,

Good news. Oxford-AstraZeneca COVID-19 Vaccine reported promising results. You might've read that we could have a vaccine by September. Experts say January 2021 is more realistic. 

THE STORIES 

⚡Nuclear can save us

☃️ Winter is coming?

ENERGY

⚡ Nuclear can save us

 

WHAT 

  • We hear it all the time: end fossil fuels. But how can we transition away from fossil fuels—which powers more than 60% of our country today—when there’s so much of it at a low price? 

  • According to Harvard Professor Stephen Pinker, the (wrong) answer many give is renewable energy. Renewables can’t run around the clock. Batteries can’t store energy for weeks, or even a few days, to power cities (although Elon is working on this). 

  • Let’s look abroad: Germany went all in for renewables. Their results so far? Little reduction in carbon emissions to date and 100+ years away from decarbonizing their grid at their current rate.

  • The (right) answer according to Pinker: Nuclear energy. France & Sweden replaced almost all their fossil fuels with nuclear energy in < 20 years. They now emit less than a tenth of the world average of carbon dioxide per kilowatt-hour. Their electricity is also cheaper than Germany’s.

WHO

  • Steven Pinker is a psychology professor at Harvard University. Through his work, he believes that if you can sway public opinion on nuclear energy through psychology, then you can solve the climate crisis.  

WHY SHOULD I CARE 

  • Why don’t we just switch to nuclear energy? Public opinion. The public doesn’t think nuclear energy is safe because of disasters such as Chernobyl. The US also doesn’t want to purchase reactors from China, who produces them at ⅙ of the cost we do.

  • Fossil fuels are driving the global warming crisis. In fact, the burning of fossil fuels accounts for ¾ of carbon emissions.

TL;DR 

  • Many advocates for green energy think the solution is renewables. Professor Pinker calls this a “fantasy” and calls for nuclear energy to solve our climate crisis. The problem? Public opinion.

READ MORE HERE
 


COMPUTER SCIENCE

☃️ Winter is coming?

 

WHAT 

  • Deep learning has fueled many recent breakthroughs in AI, such as computer vision, translation, language understanding, and more.

  • However, a recent paper argues that much of this progress comes from researchers using more computing resources to train their deep learning models rather than developing better deep learning techniques.

  • This overreliance on computing power is problematic because, as Moore’s Law nears its physical limits, we will soon no longer be able to increase computing power at the same rate as we did over the past 50 years.

  • The paper argues that unless we manage to squeeze more computing power out of traditional chips, develop innovations that make deep learning more efficient, or develop serious breakthroughs in quantum computing, progress in AI systems built on deep learning will slow dramatically.

WHO

  • Neil Thompson is a research scientist at MIT’s Computer Science and A.I. Lab (CSAIL) and MIT’s Initiative on the Digital Economy. Dr. Thompson has advised governments and businesses on the future of Moore’s Law. 

WHY SHOULD I CARE 

  • If computational limits slow the development of AI, then many of the predictions about AI will take longer than initially expected.

  • This might be good news for a long haul truck driver, but it would also mean that advances in areas like cybersecurity and vaccine development will be limited.

TL;DR 

  • Progress in areas of AI that rely on deep learning, such as image recognition and language translation, may proceed at a slower pace than it did in the past decade because deep learning models’ appetite for computing resources is growing faster than our ability to increase the computing capacity of state of the art chips.

READ MORE HERE
 

PROFESSOR'S CORNER

Wake up on the wrong side of the bed this morning? Hope not, because we got a good brainteaser today that Google interviewers used to ask. 

  • Let’s say that you have 25 horses, and you want to pick the fastest 3 horses out of those 25. In each race, only 5 horses can run at the same time because there are only 5 tracks.
What is the minimum number of races required to find the 3 fastest horses without using a stopwatch?

Recommended by Allen W., a Googler and one of our readers!
Check out the solution
Twitter
Facebook
Website
Copyright © 2020 The Office Hour, All rights reserved.


Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list.