Vitaly Vanchurin: The Universe Is a Neural Network That Learns

Theories of Everything with Curt Jaimungal

Curt Jaimungal

Physics, Philosophy, Society & Culture, Science

4.6 • 606 Ratings

🗓️ 9 February 2026

⏱️ 116 minutes

🔗️ Recording | iTunes | RSS

🧾️ Download transcript

Summary

What if physics is the universe learning? Professor Vitaly Vanchurin argues the cosmos isn't merely modeled by neural networks — it literally is one. Learning dynamics aren't a metaphor for physics; they are the physics. He calls for unifying QM, GR, and observers. I personally subscribe to The Economist. TOE listeners get 35% off the annual subscription. No other podcast has this! https://economist.com/TOE TIMESTAMPS: - 00:00:00 - The Neural Network Universe - 00:05:48 - Learning Dynamics as Physics - 00:11:52 - Optimization and Variational Principles - 00:21:17 - Deriving Fundamental Field Equations - 00:28:47 - Fermions and Particle Emergence - 00:37:17 - Geometry of Learning Algorithms - 00:44:53 - Emergent Quantum Mechanics - 00:50:01 - Renormalization and Interpretability - 00:57:00 - Second Law of Learning - 01:05:10 - Subatomic Natural Selection - 01:15:40 - Consciousness and Learning Efficiency - 01:24:09 - Unifying Physics and Observers - 01:31:01 - Qualia and Hidden Variables - 01:40:24 - Free Energy Principle Integration - 01:46:04 - Epistemological Doubt and Advice LINKS: - Vitaly's Papers: https://inspirebeta.net/literature?sort=mostrecent&size=25&page=1&q=find%20author%20vanchurin - Vitaly's Lecture: https://youtu.be/TagDLiLb2VQ - Vitaly's Website: https://cosmos.phy.tufts.edu/~vitaly/ - Towards A Theory Of Machine Learning: https://arxiv.org/pdf/2004.09280 - Autonomous Particles: https://arxiv.org/pdf/2301.10077 - Emergent Field Theories From Neural Networks: https://arxiv.org/pdf/2411.08138 - Covariant Gradient Descent: https://arxiv.org/pdf/2504.05279 - A Quantum-Classical Duality And Emergent Spacetime: https://arxiv.org/abs/1903.06083 - Emergent Quantumness In Neural Networks: https://arxiv.org/abs/2012.05082 - Predictability Crisis In Inflationary Cosmology: https://arxiv.org/abs/gr-qc/9905097 - Stationary Measure In The Multiverse: https://arxiv.org/abs/0812.0005 - The World As A Neural Network: https://arxiv.org/pdf/2008.01540 - Self-Organized Criticality In Neural Networks: https://arxiv.org/pdf/2107.03402v1 - One Hundred Authors Against Einstein: https://amazon.com/dp/B09PHH7KC8?tag=toe08-20 - Geocentric Cosmology: https://arxiv.org/abs/1006.4148 - Jacob Barandes [TOE]: https://youtu.be/gEK4-XtMwro - Yang-Hui He [TOE]: https://youtu.be/spIquD_mBFk - Eva Miranda [TOE]: https://youtu.be/6XyMepn-AZo - Felix Finster [TOE]: https://youtu.be/fXzO_KAqrh0 - Stephen Wolfram [TOE]: https://youtu.be/FkYer0xP37E - Stephen Wolfram 2 [TOE]: https://youtu.be/0YRlQQw0d-4 - Avshalom Elitzur [TOE]: https://youtu.be/pWRAaimQT1E - Ted Jacobson [TOE]: https://youtu.be/3mhctWlXyV8 - Geoffrey Hinton [TOE]: https://youtu.be/b_DUft-BdIE - Wayne Myrvold [TOE]: https://youtu.be/HIoviZe14pY - Cumrun Vafa [TOE]: https://youtu.be/kUHOoMX4Bqw - Claudia De Rham [TOE]: https://youtu.be/Ve_Mpd6dGv8 - Lee Smolin [TOE]: https://youtu.be/uOKOodQXjhc - Consciousness Iceberg [TOE]: https://youtu.be/65yjqIDghEk - Matthew Segall [TOE]: https://youtu.be/DeTm4fSXpbM - Andres Emilsson [TOE]: https://youtu.be/BBP8WZpYp0Y - Will Hahn [TOE]: https://youtu.be/3fkg0uTA3qU - David Wallace [TOE]: https://youtu.be/4MjNuJK5RzM - Karl Friston [TOE]: https://youtu.be/uk4NZorRjCo Theories of Everything with Curt Jaimungal features long-form, technically detailed interviews with leading researchers in physics, mathematics, consciousness, and philosophy, exploring topics at the level of active research. For academics, graduate students, and anyone seeking depth beyond popular science. FOLLOW: Substack | Spotify | YouTube | Twitter Learn more about your ad choices. Visit megaphone.fm/adchoices Learn more about your ad choices. Visit megaphone.fm/adchoices

Transcript

Click on a timestamp to play from that location

0:00.0	Most of my best ideas don't happen during interviews. They come spontaneously, maybe in the
0:05.8	shower or while I'm walking, and until Plaud, I kept losing them because by the time I write it
0:11.5	down, half of it's gone. I've tried voice capture before, like Google Home and just cuts me off
0:16.4	in the middle of a thought, and I don't know about you, but my ideas don't come in these 10-second short soundbites.
0:22.2	They're ponderous, they wind, they're often five minutes long, and Apple notes, Google Keep,
0:27.5	the transcription is quite horrible, and you even have to do multiple taps to get to it.
0:31.9	Plod lets me talk for as long as I want to. There's no interruptions, it's accurate capture,
0:37.0	and it organizes everything
0:38.2	into clear summaries, key takeaways, and action items. I can even come back later and say,
0:42.9	okay, what was that thread that I was talking about about consciousness and information? My personal
0:47.3	workflow is that I have their auto flow feature enabled, and it sends me an email whenever I take a note.
0:52.3	I have the note pin S for the shower,
0:54.9	and then I carry this one around me in the apartment,
0:57.4	and I love them both very much, especially this one.
1:01.6	The fact that I can just press it,
1:03.2	and it turns on instantly and starts recording without a delay,
1:06.4	is an extremely underrated feature.
1:08.7	And it's battery.
1:09.5	I haven't had to charge this since I received it.
1:12.1	Over one and a half million people use plot around the world. If your work depends on conversations
1:16.8	or the ideas that come after them, it's worth checking out. That's p-l-a-u-d.a-I-t,
1:23.2	use code T-O-E for 10% off at checkout.
	...

Please login to see the full transcript.

Previous episode | Next episode

Disclaimer: The podcast and artwork embedded on this page are from Curt Jaimungal, and are the property of its owner and not affiliated with or endorsed by Tapesearch.

Generated transcripts are the property of Curt Jaimungal and are distributed freely under the Fair Use doctrine. Transcripts generated by Tapesearch are not guaranteed to be accurate.