Sam Lehman: What the Reinforcement Learning Renaissance Means for Decentralized AI

Sam Lehman: What the Reinforcement Learning Renaissance Means for Decentralized AI

Join Tommy Shaughnessy from Delphi Ventures as he hosts Sam Lehman, Principal at Symbolic Capital and AI researcher, for a deep dive into the Reinforcement Learning (RL) renaissance and its implications for decentralized AI. Sam recently authored a widely discussed post, "The World's RL Gym", exploring the evolution of AI scaling and the exciting potential of decentralized networks for training next-generation models.

The World’s RL Gym: https://www.symbolic.capital/writing/the-worlds-rl-gym



🎯 Key Highlights


The three phases of AI scaling: Pre-training, Inference Time Compute, and the RL Renaissance.

How DeepMind's novel RL approach (using GRPO) created powerful reasoning models with minimal human data.

Understanding "reasoning traces" and how models learn to "think" longer and more effectively.

The potential downsides of human preference data potentially inhibiting model creativity, drawing parallels to AlphaGo.

Exploring the "World's RL Gym" concept: Decentralizing RL through open environments, diverse tasks, and verified data.

Why open, collaborative RL environments might outperform closed-source labs in generating diverse AI strategies.

The critical role of high-quality base models for successful RL fine-tuning.

Future AI architectures: Continuous learning and the potential of modular Mixture-of-Experts (MoE) models.

Current landscape: Open-source vs. proprietary AI, the challenge of model lock-in, and the role of crypto networks.

Debunking recent claims that "RL is dead" and understanding its true impact.



💡 Want to stay updated with the latest in crypto & AI? Hit subscribe and the notification bell! 🔔



🧠 Follow the Alpha


Tommy's Twitter: @Shaughnessy119

Sam's Twitter: @SPLehman

Symbolic Capital’s Twitter: @symbolicvc



🔗 Connect with Delphi


🌐 Portal: https://delphidigital.io/

🐦 Twitter: https://twitter.com/delphi_digital

💼 LinkedIn: https://www.linkedin.com/company/delphi-digital



🎧 Listen on


Spotify: https://open.spotify.com/show/62PR1RigLG2YN5Pelq6UY9?si=18ac7ccf36ab4753

Apple Podcasts: https://podcasts.apple.com/us/podcast/the-delphi-podcast/id1438148082

Youtube: https://www.youtube.com/channel/UC9Yy99ZlQIX9-PdG_xHj43Q



Timestamps


00:00 - Introduction: Sam Lehman, Symbolic Capital & "The World's RL Gym"

01:30 - History of AI Scaling: Pre-training Era

03:30 - Phase 2: Inference Time Compute Scaling

09:30 - Phase 3: The RL Renaissance & DeepMind Moment

14:30 - How DeepMind Trained R1 without Human Preferences

16:30 - AlphaGo Analogy: Human Data Inhibiting Creativity?

20:30 - Generalizability of RL Training: How Far Does It Go?

22:30 - The "Aha Moment": Models Learning to Think Longer

25:30 - Concept: Decentralized RL & The World's Gym

31:30 - Why Decentralize RL? Open Collaboration vs. Closed Labs

35:00 - Understanding Reasoning Traces

39:00 - Current Decentralized RL Projects (Prime Intellect, General Reasoning)

41:30 - Future Architectures: Continuous Improvement & Modular Models

46:30 - Open Source vs. Proprietary AI: Landscape & Challenges

50:30 - The Lock-In Problem with Foundational Models

52:30 - Is AGI Here? Experiences with GPT-4o

56:30 - Investment Focus in Decentralized AI

59:00 - Modular MoE Models & Jensen's HDEE Paper

1:03:00 - Debunking "RL is Dead" Claims

1:06:00 - Importance of Performant Base Models for RL



Disclaimer


This podcast is strictly informational and educational and is not investment advice or a solicitation to buy or sell any tokens or securities or to make any financial decisions. Do not trade or invest in any project, tokens, or securities based upon this podcast episode. The host and members at Delphi Ventures may personally own tokens or art that are mentioned on the podcast. Our current show features paid sponsorships which may be featured at the start, middle, and/or the end of the episode. These sponsorships are for informational purposes only and are not a solicitation to use any product, service or token.

Avsnitt(468)

Storecoin’s Chris and Rag: On A Radically New Zero-Fee, P2P Cloud Computing Platform

Storecoin’s Chris and Rag: On A Radically New Zero-Fee, P2P Cloud Computing Platform

In this episode, Host Tom Shaughnessy of Delphi Digital (www.DelphiDigital.io) is joined by Chris McCoy the co-founder of Storecoin and co-founder and CTO Rag Bhagavatha. The conversation covers the new settlements platform layer Storecoin is building and the co-founder's goals of also building a cloud layer on-top of this layer. Its a differentiated and novel episode and worth a listen.  To ensure you’re able to submit questions moving forward and receive our analysts calls like clock-work, visit DelphiDigital.io on your device and sign up now.   Follow Tom on Twitter @Shaughnessy119 Follow Chris on Twitter @chrisamccoy  Follow Storecoin on Twitter @storecoin   Disclosure: Tom Shaughnessy owns tokens in ETH, DCR, MKR, ZRX and HYDRO. This podcast is NOT investment advice and is only informational. Do not make investment decisions based upon this podcast. Delphi Digital was not compensated by any party for this podcast episode other than Podbean's advertisers. This content is strictly informational.  - Advertisers: To advertise on this podcast, email Tom@DelphiDigital.io Potential Guests: If you're interested in appearing on the podcast, email Tom@DelphiDigital.io  -

21 Mars 20191h 12min

Delphi Digital’s March Analyst Call - Ethereum, Enjin and Our Short Term Bitcoin Outlook

Delphi Digital’s March Analyst Call - Ethereum, Enjin and Our Short Term Bitcoin Outlook

In this special episode, the entire Delphi Digital (www.DelphiDigital.io) analyst team shares our monthly analyst call, which is usually for institutional clients only. We dive into our massive Ethereum report, an update on Enjin which we released, a breakdown of smart contract competition following the launch of Cosmos and so much more. This call happens monthly for our institutional clients, will be for members only moving forward, and includes all 5 analysts at Delphi Digital; Tom Shaughnessy, Kevin Kelly, Yan Liberman, Anil Lulla, and Medio Demarco. To ensure you’re able to submit questions moving forward and receive our analysts calls like clock-work, visit DelphiDigital.io on your device and sign up now.   Follow Tom on Twitter @Shaughnessy119 Follow Kevin on Twitter @Kevin_Kelly_II  Follow Yan on Twitter @YanLiberman  Follow Anil on Twitter @anildelphi  Follow Medio on Twitter @mediodelphi    Disclosure: Tom Shaughnessy owns tokens in ETH, DCR, MKR, ZRX and HYDRO. This podcast is NOT investment advice and is only informational. Do not make investment decisions based upon this podcast. Delphi Digital was not compensated by any party for this podcast episode other than Podbean's advertisers. This content is strictly informational.  - Advertisers: To advertise on this podcast, email Tom@DelphiDigital.io Potential Guests: If you're interested in appearing on the podcast, email Tom@DelphiDigital.io  -

19 Mars 201943min

BlockFi’s Zac Prince: Building A Fintech Platform For The Crypto Market

BlockFi’s Zac Prince: Building A Fintech Platform For The Crypto Market

Host Tom Shaughnessy of Delphi Digital (www.DelphiDigital.io) is joined by Zac Prince, The Co-Founder and CEO of BlockFi. BlockFi provides basic financial products in the blockchain ecosystem including high-interest crypto accounts and low cost credit products to clients globally.     Follow Tom on Twitter @Shaughnessy119 Follow Zac on Twitter @BlockFiZac    Disclosure: Tom Shaughnessy owns tokens in ETH, DCR, MKR, ZRX and HYDRO. This podcast is NOT investment advice and is only informational. Do not make investment decisions based upon this podcast. Delphi Digital was not compensated by any party for this podcast episode other than Podbean's advertisers. This content is strictly informational.  - Advertisers: To advertise on this podcast, email Tom@DelphiDigital.io Potential Guests: If you're interested in appearing on the podcast, email Tom@DelphiDigital.io  -

11 Mars 201950min

Max Bronstein from Dharma: #DeFi and The Importance of Debt Markets In Crypto

Max Bronstein from Dharma: #DeFi and The Importance of Debt Markets In Crypto

Host Tom Shaughnessy of Delphi Digital (www.DelphiDigital.io) is joined by Max Bronstein of the Dharma Protocol. Dharma enables users to instantly borrow and lend crypto-assets in high volume. This is a two-part DeFi series, stay tuned for part 2 next week.    Follow Tom on Twitter @Shaughnessy119 Follow Max on Twitter @max_bronstein    Disclosure: Tom Shaughnessy owns tokens in ETH, DCR, MKR, ZRX and HYDRO. This podcast is NOT investment advice and is only informational. Do not make investment decisions based upon this podcast. Delphi Digital was not compensated by any party for this podcast episode other than Podbean's advertisers. This content is strictly informational.  - Advertisers: To advertise on this podcast, email Tom@DelphiDigital.io Potential Guests: If you're interested in appearing on the podcast, email Tom@DelphiDigital.io  -

1 Mars 201957min

Crypto Roundtable: Tom Shaughnessy, Gregory Rocco, Tanner Hoban and Sam Corso

Crypto Roundtable: Tom Shaughnessy, Gregory Rocco, Tanner Hoban and Sam Corso

Host Tom Shaughnessy of Delphi Digital (www.DelphiDigital.io) is joined by Gregory Rocco of ConsenSys' Alpine group. Tanner Hoban of ConsenSys Digital securities and Sam Corso who runs Techsuite.io and is an advisor to Delphi digital.   Follow Tom on Twitter @Shaughnessy119 Follow Rocco on Twitter @Obstropolos  Follow Tanner on Twitter @tehoban1  Follow Sam on Twitter @samc621    Disclosure: Tom Shaughnessy owns tokens in ETH, DCR, MKR, ZRX and HYDRO. This podcast is NOT investment advice and is only informational. Do not make investment decisions based upon this podcast. Delphi Digital was not compensated by any party for this podcast episode other than Podbean's advertisers. This content is strictly informational.  - Advertisers: To advertise on this podcast, email Tom@DelphiDigital.io Potential Guests: If you're interested in appearing on the podcast, email Tom@DelphiDigital.io  -

25 Feb 20191h 26min

Nathaniel Whittemore: The Storyteller Is King In The Land of Crypto Sentiment

Nathaniel Whittemore: The Storyteller Is King In The Land of Crypto Sentiment

Host Tom Shaughnessy of Delphi Digital (www.DelphiDigital.io) is joined by Nathaniel Whittemore Nathaniel is in my opinion not only the best story-teller in the space, but has the best ability at synthesizing down the intent and goal of projects into digestible and understandable words. If someone was explaining rocket science I would want it to be Nathaniel. Follow Tom on Twitter @Shaughnessy119 Follow Nathaniel on Twitter @nlw    Disclosure: Tom Shaughnessy owns tokens in ETH, DCR, MKR, ZRX, HYDRO, and GLA. This podcast is NOT investment advice and is only informational. Do not make investment decisions based upon this podcast. Delphi Digital was not compensated by any party for this podcast episode other than Podbean's advertisers. This content is strictly informational.  - Advertisers: To advertise on this podcast, email Tom@DelphiDigital.io Potential Guests: If you're interested in appearing on the podcast, email Tom@DelphiDigital.io  -

18 Feb 20191h 10min

Jack Platts: Web3, Polkadot and The Promise Of Interoperability

Jack Platts: Web3, Polkadot and The Promise Of Interoperability

Host Tom Shaughnessy of 51percent Crypto Research (www.51pct.io) is joined by the Web3 Foundation's Jack Platts. Jack covers the Web3 foundation; and we focus the majority of the conversation on Polkadot which is connected to Web3. Polkadot is a major interoperability protocol set to launch and we cover the mechanics of the protocol, security, parachains, competition (Cosmos, Ethereum et al) and so much more.   Follow Tom on Twitter @Shaughnessy119 Follow Jack on Twitter @web3jp  - Related Posts:   Interoperability and Composability: Killer of Redundant Blockchains (Part 1) Polkadot vs Cosmos: The Disastrous Issue and Crowning A Winner (Part 2) - 51percent's Institutional Crypto Podcasts are to the point discussions with crypto leaders for analysts, funds and institutions. Make sure to add your email on 51pct.io Disclosure: Tom Shaughnessy owns tokens in ETH, DCR, MKR, ZRX, HYDRO, and GLA. This podcast is NOT investment advice and is only informational. Do not make investment decisions based upon this podcast. 51percent was not compensated by any party for this podcast episode. Tom has no investment stake in Parity. This content is strictly informational.   Sign Up For 51percent's Leading Crypto Research  - Advertisers: To advertise on this podcast, email Tom@51pct.io Potential Guests: If you're interested in appearing on the podcast, email Tom@51pct.io  -

11 Feb 201949min

TokenSoft’s CEO Mason Borda: Blockchain Technology Enabling a 24/7 and Global Digital Security Market

TokenSoft’s CEO Mason Borda: Blockchain Technology Enabling a 24/7 and Global Digital Security Market

Host Tom Shaughnessy of 51percent Crypto Research (www.51pct.io) is joined by Tokensoft's co-founder and CEO Mason Borda TokenSoft enables small businesses, enterprises, and financial institutions to meet compliance requirements at issuance, distribution or exchange. This is a great episode as we have already had on numerous STO companies, so Tokensoft rounds out our STO coverage! Follow Tom on Twitter @Shaughnessy119 Follow Mason on Twitter @masonic_tweets   - 51percent's Institutional Crypto Podcasts are to the point discussions with crypto leaders for analysts, funds and institutions. Make sure to add your email on 51pct.io Disclosure: Tom Shaughnessy owns tokens in ETH, MKR, ZRX, HYDRO, CVC, DCR, POLY and GLA. This podcast is NOT investment advice and is only informational. Do not make investment decisions based upon this podcast. 51percent was not compensated by any party for this podcast episode. Tom has no investment stake in Parity. This content is strictly informational.   Sign Up For 51percent's Leading Crypto Research  - Advertisers: To advertise on this podcast, email Tom@51pct.io Potential Guests: If you're interested in appearing on the podcast, email Tom@51pct.io  -

4 Feb 201955min

Populärt inom Teknik

uppgang-och-fall
elbilsveckan
market-makers
rss-racevecka
rss-elektrikerpodden
rss-uppgang-och-fall
natets-morka-sida
bli-saker-podden
skogsforum-podcast
rss-technokratin
developers-mer-an-bara-kod
har-vi-akt-till-mars-an
solcellskollens-podcast
rss-veckans-ai
mediepodden
bilar-med-sladd
rss-laddstationen-med-elbilen-i-sverige
hej-bruksbil
rss-fabriken-2
rss-bakom-boken