Artwork

Content provided by The Technium. All podcast content including episodes, graphics, and podcast descriptions are uploaded and provided directly by The Technium or their podcast platform partner. If you believe someone is using your copyrighted work without your permission, you can follow the process outlined here https://player.fm/legal.
Player FM - Podcast App
Go offline with the Player FM app!

Multi-modal and Multi-task Models (S03 E04)

1:13:33
 
Share
 

Manage episode 334037654 series 3314588
Content provided by The Technium. All podcast content including episodes, graphics, and podcast descriptions are uploaded and provided directly by The Technium or their podcast platform partner. If you believe someone is using your copyrighted work without your permission, you can follow the process outlined here https://player.fm/legal.

Multimodal and multitask models are machine learning models that can generalize. Multimodal models can generalize to understand different types of input, for example images and text. Multitask models can generalize their knowledge by applying what they’ve learned about one task to solve another task.

Links/Resources:
• MUM: https://blog.google/products/search/introducing-mum/
• Gato: https://www.youtube.com/watch?v=wSQJZHfAg18
• MIA: https://www.youtube.com/watch?v=L9kA8nSJdYw
• Flamingo: https://www.deepmind.com/blog/tackling-multiple-tasks-with-a-single-visual-language-model
• Flamingo explaining a funny photo: https://twitter.com/MelMitchell1/status/1522642194741538817
• Is LaMDA Sentient?: https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917

Chapters:
0:00 Intros
2:33 Multimodal and Multitasks Models
6:50 Deepmind's Gato: The All-Rounder Athlete
14:43 Google's MUM: The Search Assistant
18:12 Deepmind's Multimodal Interactive Agent: The Domestic Helper
22:31 Deepmind's Flamingo: Reasoning about Pictures
26:45 Why are these mind-blowing?
31:20 Machine Learning has come a looooong way
35:21 Could Flamingo be the real JARVIS?
38:56 Could MIA assist the elderly?
43:05 Multimodal AI for self driving cars
51:15 Multitask = A Shared Brain That Learns Everything
1:00:19 Could these models transcend human knowledge?
1:08:50 Breaking news: AI models are sentient
1:10:37 Is this just a local maximum or a path to AGI?
1:11:50 Outros

===== About “The Technium” =====
The Technium is a weekly podcast discussing the edge of technology and what we can build with it. Each week, Sri and Wil introduce a big idea in the future of computing and extrapolate the effect it will have on the world.
Follow us for new videos every week on web3, cryptocurrency, programming languages, machine learning, artificial intelligence, and more!

===== Socials =====
WEBSITE: https://technium.transistor.fm/
SPOTIFY: https://open.spotify.com/show/1ljTFMgTeRQJ69KRWAkBy7
APPLE PODCASTS: https://podcasts.apple.com/us/podcast/the-technium/id1608747545

  continue reading

40 episodes

Artwork
iconShare
 
Manage episode 334037654 series 3314588
Content provided by The Technium. All podcast content including episodes, graphics, and podcast descriptions are uploaded and provided directly by The Technium or their podcast platform partner. If you believe someone is using your copyrighted work without your permission, you can follow the process outlined here https://player.fm/legal.

Multimodal and multitask models are machine learning models that can generalize. Multimodal models can generalize to understand different types of input, for example images and text. Multitask models can generalize their knowledge by applying what they’ve learned about one task to solve another task.

Links/Resources:
• MUM: https://blog.google/products/search/introducing-mum/
• Gato: https://www.youtube.com/watch?v=wSQJZHfAg18
• MIA: https://www.youtube.com/watch?v=L9kA8nSJdYw
• Flamingo: https://www.deepmind.com/blog/tackling-multiple-tasks-with-a-single-visual-language-model
• Flamingo explaining a funny photo: https://twitter.com/MelMitchell1/status/1522642194741538817
• Is LaMDA Sentient?: https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917

Chapters:
0:00 Intros
2:33 Multimodal and Multitasks Models
6:50 Deepmind's Gato: The All-Rounder Athlete
14:43 Google's MUM: The Search Assistant
18:12 Deepmind's Multimodal Interactive Agent: The Domestic Helper
22:31 Deepmind's Flamingo: Reasoning about Pictures
26:45 Why are these mind-blowing?
31:20 Machine Learning has come a looooong way
35:21 Could Flamingo be the real JARVIS?
38:56 Could MIA assist the elderly?
43:05 Multimodal AI for self driving cars
51:15 Multitask = A Shared Brain That Learns Everything
1:00:19 Could these models transcend human knowledge?
1:08:50 Breaking news: AI models are sentient
1:10:37 Is this just a local maximum or a path to AGI?
1:11:50 Outros

===== About “The Technium” =====
The Technium is a weekly podcast discussing the edge of technology and what we can build with it. Each week, Sri and Wil introduce a big idea in the future of computing and extrapolate the effect it will have on the world.
Follow us for new videos every week on web3, cryptocurrency, programming languages, machine learning, artificial intelligence, and more!

===== Socials =====
WEBSITE: https://technium.transistor.fm/
SPOTIFY: https://open.spotify.com/show/1ljTFMgTeRQJ69KRWAkBy7
APPLE PODCASTS: https://podcasts.apple.com/us/podcast/the-technium/id1608747545

  continue reading

40 episodes

All episodes

×
 
Loading …

Welcome to Player FM!

Player FM is scanning the web for high-quality podcasts for you to enjoy right now. It's the best podcast app and works on Android, iPhone, and the web. Signup to sync subscriptions across devices.

 

Quick Reference Guide