#108 - Dr. JOEL LEHMAN - Machine Love [Staff Favourite]

#108 - Dr. JOEL LEHMAN - Machine Love [Staff Favourite]

Support us! https://www.patreon.com/mlst

MLST Discord: https://discord.gg/aNPkGUQtc5


We are honoured to welcome Dr. Joel Lehman, an eminent machine learning research scientist, whose work in AI safety, reinforcement learning, creative open-ended search algorithms, and indeed the philosophy of open-endedness and abandoning objectives has paved the way for innovative ideas that challenge our preconceptions and inspire new visions for the future.

Dr. Lehman's thought-provoking book, "Why Greatness Cannot Be Planned" penned with with our MLST favourite Professor Kenneth Stanley has left an indelible mark on the field and profoundly impacted the way we view innovation and the serendipitous nature of discovery. Those of you who haven't watched our special edition show on that, should do so at your earliest convenience! Building upon this foundation, Dr. Lehman has ventured into the domain of AI systems that embody principles of love, care, responsibility, respect, and knowledge, drawing from the works of Maslow, Erich Fromm, and positive psychology.


YT version: https://youtu.be/23-TXgJEv-Q


http://joellehman.com/

https://twitter.com/joelbot3000


Interviewer: Dr. Tim Scarfe


TOC:

Intro [00:00:00]

Model [00:04:26]

Intro and Paper Intro [00:08:52]

Subjectivity [00:16:07]

Reflections on Greatness Book [00:19:30]

Representing Subjectivity [00:29:24]

Nagal's Bat [00:31:49]

Abstraction [00:38:58]

Love as Action Rather Than Feeling [00:42:58]

Reontologisation [00:57:38]

Self Help [01:04:15]

Meditation [01:09:02]

The Human Reward Function / Effective... [01:16:52]

Machine Hate [01:28:32]

Societal Harms [01:31:41]

Lenses We Use Obscuring Reality [01:56:36]

Meta Optimisation and Evolution [02:03:14]

Conclusion [02:07:06]


References:


What Is It Like to Be a Bat? (Thomas Nagel)

https://warwick.ac.uk/fac/cross_fac/iatl/study/ugmodules/humananimalstudies/lectures/32/nagel_bat.pdf


Why Greatness Cannot Be Planned: The Myth of the Objective (Kenneth O. Stanley and Joel Lehman)

https://link.springer.com/book/10.1007/978-3-319-15524-1


Machine Love (Joel Lehman)

https://arxiv.org/abs/2302.09248


How effective altruists ignored risk (Carla Cremer)

https://www.vox.com/future-perfect/23569519/effective-altrusim-sam-bankman-fried-will-macaskill-ea-risk-decentralization-philanthropy


Philosophy tube - The Rich Have Their Own Ethics: Effective Altruism

https://www.youtube.com/watch?v=Lm0vHQYKI-Y


Abandoning Objectives: Evolution through the Search for Novelty Alone (Joel Lehman and Kenneth O. Stanley)

https://www.cs.swarthmore.edu/~meeden/DevelopmentalRobotics/lehman_ecj11.pdf

Denne episoden er hentet fra en åpen RSS-feed og er ikke publisert av Podme. Den kan derfor inneholde annonser.

Episoder(271)

What Most People Get Wrong About Evolution | Akarsh Kumar

What Most People Get Wrong About Evolution | Akarsh Kumar

A lot of people in AI treat evolution as a dumb fallback, basically random search for when you can't take a gradient. Akarsh Kumar thinks that is wrong. Selection hangs on to partial solutions, so mut...

10 Okt 47min

How a Voice Agent Learns the Rhythm of Conversation — Shawn Wen

How a Voice Agent Learns the Rhythm of Conversation — Shawn Wen

Tsung-Hsien (Shawn) Wen, CTO of PolyAI, tells Tim Scarfe why voice agents are harder than text agents. Voice adds time, and a good conversation depends on adapting to the person on the line, not just ...

1 Okt 1h 10min

Who Checks a Proof No Human Can Read? — Leo de Moura

Who Checks a Proof No Human Can Read? — Leo de Moura

Leonardo de Moura created Lean and co-created Z3. ---This episode is sponsored by Parallel.Parallel, where agents find answers: web search, extraction and deep research APIs built for AI agents.Start ...

30 Sep 1h 14min

When AI Research Starts Moving Faster Than Human Research - Zhengyao Jiang

When AI Research Starts Moving Faster Than Human Research - Zhengyao Jiang

Weco let an AI coding agent rewrite the harness around another agent for eight days: its code, prompts and tools, while the underlying language model stayed fixed. Tim Scarfe asks Weco co-founder Zhen...

26 Sep 43min

How Deep Learning Finally Cracked Messy Tables - Frank Hutter

How Deep Learning Finally Cracked Messy Tables - Frank Hutter

Frank Hutter, co-founder of Prior Labs, talks about TabPFN, a tabular foundation model that makes predictions in a single forward pass, and the research behind it.TabPFN is pre-trained on synthetic da...

23 Sep 1h 53min

Why Scaling Prediction Cannot Create Intelligence - Alexander Mattick

Why Scaling Prediction Cannot Create Intelligence - Alexander Mattick

Alexander Mattick is a researcher at Fraunhofer IIS and a PhD researcher at the University of Technology Nuremberg (UTN), and a regular on Yannic Kilcher's Discord. He first came on MLST in 2022, afte...

21 Sep 2h 14min

How Physical AI Learns Across Language, Video and Action — Ming-Yu Liu

How Physical AI Learns Across Language, Video and Action — Ming-Yu Liu

The car making a left turn at the start of this episode was never filmed. Cosmos 3 generated it. Ming-Yu Liu, who leads the Cosmos research at NVIDIA, explains how one model can describe a video, gene...

15 Sep 25min

Speech Recognition Is Not a Solved Problem — Pavan Kumar Reddy

Speech Recognition Is Not a Solved Problem — Pavan Kumar Reddy

Pavan Kumar Reddy leads audio research at Mistral AI. He joins Tim Scarfe for a deep technical tour of Voxtral — and explains why the frontier of deployed voice is still a cascade of specialised model...

14 Sep 1h 42min

Populært innen Teknologi

teknisk-sett
energi-og-klima
elektropodden
lydartikler-fra-aftenposten
nasjonal-sikkerhetsmyndighet-nsm
tomprat-med-gunnar-tjomlid
shifter
rss-ai-forklart
rss-alt-som-gar-pa-strom
smart-forklart
rss-kunstig-intelligens-med-elisabeth-maren-og-morten
fornybaren
rss-anleggspraten
pedagogisk-intelligens
kortslutning
rss-hvorfor-ble-det-sann
rss-alt-vi-kan
rss-larervarelset
rss-romnasjonen
i-loopen