Lex Fridman Podcast - #222 – Jay McClelland: Neural Networks and the Emergence of Cognition
Jay McClelland is a cognitive scientist at Stanford. Please support this podcast by checking out our sponsors:
– Paperspace: https://gradient.run/lex to get $15 credit
– Skiff: https://skiff.org/lex to get early access
– Uprising Food: https://uprisingfood.com/lex to get $10 off 1st starter bundle
– Four Sigmatic: https://foursigmatic.com/lex and use code LexPod to get up to 60% off
– Onnit: https://lexfridman.com/onnit to get up to 10% off
EPISODE LINKS:
Jay’s Website: https://stanford.edu/~jlmcc/
PODCAST INFO:
Podcast website: https://lexfridman.com/podcast
Apple Podcasts: https://apple.co/2lwqZIr
Spotify: https://spoti.fi/2nEwCF8
RSS: https://lexfridman.com/feed/podcast/
YouTube Full Episodes: https://youtube.com/lexfridman
YouTube Clips: https://youtube.com/lexclips
SUPPORT & CONNECT:
– Check out the sponsors above, it’s the best way to support this podcast
– Support on Patreon: https://www.patreon.com/lexfridman
– Twitter: https://twitter.com/lexfridman
– Instagram: https://www.instagram.com/lexfridman
– LinkedIn: https://www.linkedin.com/in/lexfridman
– Facebook: https://www.facebook.com/lexfridman
– Medium: https://medium.com/@lexfridman
OUTLINE:
Here’s the timestamps for the episode. On some podcast players you should be able to click the timestamp to jump to that time.
(00:00) – Introduction
(07:12) – Beauty in neural networks
(11:31) – Darwin and evolution
(17:16) – The origin of intelligence
(23:58) – Explorations in cognition
(30:02) – Learning representations by back-propagating errors
(36:27) – Dave Rumelhart and cognitive modeling
(49:30) – Connectionism
(1:12:23) – Geoffrey Hinton
(1:14:19) – Learning in a neural network
(1:31:11) – Mathematics & reality
(1:38:19) – Modeling intelligence
(1:48:57) – Noam Chomsky and linguistic cognition
(2:03:18) – Advice for young people
(2:14:26) – Psychiatry and exploring the mind
(2:27:04) – Legacy
(2:32:53) – Meaning of life
Big Technology Podcast - A Look Into Facebook’s Soul — With WSJ’s Jeff Horwitz and Ex-FB Exec. Brian Boland
Jeff Horwitz is the Wall Street Journal reporter who unearthed a trove of internal Facebook documents that reveal a damning disconnect between what the company says in public and its actions inside. Brian Boland is a former Facebook executive who spent more than 11 years inside the company. The two come on to discuss Horowitz's bombshell series of reports, unpacking what they tell us about Facebook, dissecting the company's responses, and looking at potential solutions.
Enjoy this bonus episode!
Talk Python To Me - #334: Microsoft Planetary Computer
PHPUgly - 254:Testing Patience
Links from the show:
- Write Unit Tests Like Scenarios - Matthias Noback
- mrsuh/php-generics
- MimeType testing
- Critical vulnerabilities in JSON Web Token libraries
- Travis CI Issue
- Framework Laptops
- Linus Tech Tips Review of teh Famework Laptops
- Skipper is a powerful visual editor for Laravel Eloquent
This episode of PHPUgly was sponsored by:
- Honeybadger.io - https://www.honeybadger.io/
PHPUgly streams the recording of this podcast live. Typically every Thursday night around 9 PM PT. Come and join us, and subscribe to our Youtube Channel, Twitch, or Periscope. Also, be sure to check out our Patreon Page.
Twitter Account https://twitter.com/phpugly
Host:
Streams:
Powered by Restream
PHPUgly Anthem by Harry Mack / Harry Mack Youtube Channel
Thanks to all of our Patreon Sponsors:
Honeybadger ** This weeks Sponsor **
ButteryCrumpet
Shawn
David Q
Ken F
Tony L
Frank W
Jeff K
Shelby C
S Ferguson
Boštjan O
Matt L
Dmitri G
Knut E B
Marcus
MikePageDev
Rodrigo C
Billy
Darryl H
Blaž O
Mike W
Holly S
Peter A
Ben R
Luciano N
Elgimbo
Wayne
Kevin Y
Alex B
Clayton S
Kenrick B
R. C. S.
ahinkle
dreamup
Enno R
Sevi
Maciej P
Jeroen F
Ronny M N
Chris C
Tristan I
F'n Steve
Robert
The Stack Overflow Podcast - Writing the roadmap from engineer to manager
Former co-host Sara Chipps now manages engineering teams at LinkedIn, but her best content is still on Twitter.
Cassidy's former boss, Sarah Drasner, recently wrote a book to help engineers level up to management: Engineering Management for the Rest of Us.
Cassidy's new favorite software tool is Astro, a single-site generator that looks to minimize the amount of client-side JavaScript in a site.
The two books Ms. Chipps mention as the old standbys for new engineering managers are Peopleware and Smart and Gets Things Done.
Code Story: Insights from Startup Tech Leaders - S5 Bonus: Cody Candee, Bounce
Travel is a huge part of Code Candee's life, even prior to his current venture. He has lived in a dozen different cities, travelled to over 60 countries, and has been to 49 of the 50 US states. Out of all of them, his favorite place to live was India, specifically in Bangalore. While he was there, he tried to experience all the cultures, food and people, and quite enjoyed his time.
He is a minimalist.. to the tune of owning a couple of suitcases. He loves the idea of not being held down by things, and being able to move around at the drop of a hat. He reads a lot of books, his favorite one being the Alchemist, and enjoys rock climbing. But to be honest, he likes the all consuming, work centric lifestyle he leads.
In 2014, he was working in San Francisco and planned to have some drinks with his friends. One of them had to go all the way home to drop off some luggage, before heading to happy hour. He took the ideas he wrote down that night, and in 2017, he set out to solve this complex problem... by just getting started.
This is the creation story of Bounce.
Sponsors
- Courier
- Img.ly
- Routable
- CTO.ai
- Cloudways offers peace of mind and flexibility so you can focus on growing your business instead of dealing with server management. With Cloudways, you get an optimized stack, managed servers, backups, staging environment, integrated Git, pre-configured, Composer, 24/7 support, and a choice of five cloud providers: AWS, DigitalOcean, Linode, Google Cloud, and Vultr. Get up to 2 Month Free Hosting by using code "CODE30" and get $30 free hosting credit.
Links
- Website: https://usebounce.com/
- LinkedIn: https://www.linkedin.com/in/codycandee/
Our Sponsors:
* Check out Vanta: https://vanta.com/CODESTORY
Support this podcast at — https://redcircle.com/code-story/donations
Advertising Inquiries: https://redcircle.com/brands
Privacy & Opt-Out: https://redcircle.com/privacy
Lex Fridman Podcast - #221 – Douglas Lenat: Cyc and the Quest to Solve Common Sense Reasoning in AI
Douglas Lenat is the founder of Cyc, a 37 year project aiming to solve common-sense knowledge and reasoning in AI. Please support this podcast by checking out our sponsors:
– Squarespace: https://lexfridman.com/squarespace and use code LEX to get 10% off
– BiOptimizers: http://www.magbreakthrough.com/lex to get 10% off
– Stamps.com: https://stamps.com and use code LEX to get free postage & scale
– LMNT: https://drinkLMNT.com/lex to get free sample pack
– ExpressVPN: https://expressvpn.com/lexpod and use code LexPod to get 3 months free
EPISODE LINKS:
Douglas’s Twitter: https://twitter.com/cycorpai
Cyc’s Website: https://cyc.com
PODCAST INFO:
Podcast website: https://lexfridman.com/podcast
Apple Podcasts: https://apple.co/2lwqZIr
Spotify: https://spoti.fi/2nEwCF8
RSS: https://lexfridman.com/feed/podcast/
YouTube Full Episodes: https://youtube.com/lexfridman
YouTube Clips: https://youtube.com/lexclips
SUPPORT & CONNECT:
– Check out the sponsors above, it’s the best way to support this podcast
– Support on Patreon: https://www.patreon.com/lexfridman
– Twitter: https://twitter.com/lexfridman
– Instagram: https://www.instagram.com/lexfridman
– LinkedIn: https://www.linkedin.com/in/lexfridman
– Facebook: https://www.facebook.com/lexfridman
– Medium: https://medium.com/@lexfridman
OUTLINE:
Here’s the timestamps for the episode. On some podcast players you should be able to click the timestamp to jump to that time.
(00:00) – Introduction
(07:39) – What is Cyc?
(15:45) – How to form a knowledge base of the universe
(26:11) – How to train an AI knowledge base
(30:32) – Global consistency versus local consistency
(54:53) – Automated reasoning
(1:00:33) – Direct uses of AI and machine learning
(1:13:11) – The semantic web
(1:23:44) – Tools to help Cyc interpret data
(1:32:54) – The most beautiful idea about Cyc
(1:38:53) – Love and consciousness in AI
(1:45:52) – The greatness of Marvin Minsky
(1:50:46) – Is Cyc just a beautiful dream?
(1:55:31) – What is OpenCyc and how was it born?
(2:01:21) – The open source community and OpenCyc
(2:11:48) – The inference problem
(2:13:31) – Cyc’s programming language
(2:21:05) – Ontological engineering
(2:28:30) – Do machines think?
(2:37:15) – Death and consciousness
(2:47:16) – What would you say to AI?
(2:51:52) – Advice to young people
(2:53:48) – Mortality
The Stack Overflow Podcast - This AI-assisted bug bash is offering serious prizes for squashing nasty code
While every developer loves a good story about discovering and fixing a gnarly bug, not everyone enjoys the work of finding those bugs. Most folks would prefer to be writing business logic and solving new problems. But those input validation errors and resource leaks won’t solve themselves.
Or will they?
AWS Bug Bust is a global competition launched with the goal of finding and fixing one million bugs in codebases around the world. It takes the traditional bug bash and turns it into a competition that anyone can enter. Got a repo or two that you’ve been meaning to clean up? Enter the Bug Bust and start squashing.
This competition awards points to organizations, as well as individuals within an organization, for every bug that they fix in their own repos. A little friendly competition can motivate developers to fix more bugs in order to move up the leaderboards. How do you think we built Stack Overflow? Fake internet points are very important around here. With the Bug Bust competition, it’s not just fake internet points and personal glory; top bug squashers—overall and within top organizations—can win all expense paid trips to re:Invent 2021.
In a traditional bug bust, someone has to find the bugs, file tickets on all of them, then collect them for squashing. In the Bug Bust, Amazon has managed to automate that part of the process. That’s because the Bug Bust is built on their AI-powered code review and profiling tool, CodeGuru.
CodeGuru uses static analysis and machine learning with some additional automated reasoning to find bugs in code; everything from best practices to concurrency issues, resource leaks, security problems, and more. AI isn’t here to take your jobs, it’s here to automated away the tedious stuff. Developers get to harness the power of artificial intelligence in their everyday lives.
Concurrency and resource leak issues tend to drain the soul out of the developers. You could spend all day trying to optimize and close those. CodeGuru includes a function profiler that looks for a codebase’s most expensive calls. It’s a lightweight agent actively running and looking for ways to reduce the cost of the running application.
These bugs, along with security issues and AWS API calls, are the ones that earn the most points. But all bugs earn their bashers points; CodeGuru spots code inefficiencies, duplications, and general code quality detectors, and performs input validation. The model behind this is pretrained on years of Amazon bug hunting experience. The system does learn from you as to what is a good bug in your codebase, but it’s not training on your code. It’s your feedback that makes CodeGuru a better bug hunter.
If you have Java and Python code in a GitHub, GitHub Enterprise, Bitbucket, or AWS CodeCommit repository, you can jump into the competition. Sign up with your email and you get 30 days to run as many Bug Busts as you want for free. The top ten individual bug busters get VIP treatment at the 2021 re:Invent conference (and an all-expense-paid trip there), which is being held in person this year. Top participating organizations get a ticket to give to one of their developers as well. For those bashers outside of the top ten, you can still earn some sweet swag by passing some point milestones.
The contest to win the trip to re:Invent 2021 runs through September, but you can still automate your bug bashes and get swag anytime. Want to get started? Head over to the AWS Bug Bust site now.
Big Technology Podcast - Inside The Theranos Trial — With Erin Griffith of The New York Times
Erin Griffith is the New York Times reporter at the trial for Theranos founder Elizabeth Holmes. She joins Big Technology Podcast to bring us inside the courtroom, explaining why Holmes is on trial and whether she'll be a rare founder to face consequences for misleading investors. We also discuss whether Holmes is emblematic of the venture capital world's downsides, or an outlier.
You can find Erin on Twitter, @eringriffith
