Yearly
Weekly
Daily
Reading list
Links 25
Score: 0.9732105387750054
User feedback: None
Out links: 5092353 Raw text: 5092353Title: A conversation about Katja's counterarguments to AI risk — AI Alignment Forum Description: This post is a transcript of a conversation between Ege Erdil and Ronny Fernandez, recorded by me. The participants talked about a recent post by Kat… Keywords: No keywords Text content: A conversation ...
Score: 0.9200652406474674
User feedback: None
Out links: 5092370 Raw text: 5092370Title: AXRP Episode 31 - Singular Learning Theory with Daniel Murfet — AI Alignment Forum Description: YouTube link • What’s going on with deep learning? What sorts of models get learned, and what are the learning dynamics? Singular learning theory is… Keywords: No keywords Text content: AXRP Episod...
Score: 0.9156188206792982
User feedback: None
Out links: 290926 Raw text: 290926https://arxiv.org/pdf/2302.13861.pdf
Differentially Private Diffusion Models Generate Useful Synthetic Images Sahra Ghalebikesabi1,+ , Leonard Berrada2 , Sven Gowal2 , Ira Ktena2 , Robert Stanforth2 , Jamie Hayes2 , Soham De2 , Samuel L. Smith2 , Olivia Wiles2 and Borja Balle2 arXiv:2302.13861v1 [cs.LG] 27 Feb 2023 1 University of Ox...
Score: 0.9146148460355693
User feedback: None
Out links: 5217330 Raw text: 5217330https://alignmentforum.org/posts/LY7rovMiJ4FhHxmH5/thoughts-on-hardware-compute-requirements-for-agi
Title: Thoughts on hardware / compute requirements for AGI — AI Alignment Forum Description: [NOTE: I have some updates / corrigenda at the bottom. ] … Keywords: No keywords Text content: Thoughts on hardware / compute requirements for AGI — AI Alignment Forum This website requires javascript to p...
Score: 0.902011208960867
User feedback: None
Out links: 291713 Raw text: 291713https://arxiv.org/pdf/2404.02258.pdf
Mixture-of-Depths: Dynamically allocating compute in transformer-based language models David Raposo1* , Sam Ritter1 , Blake Richards1,2 , Timothy Lillicrap1 , Peter Conway Humphreys1 and Adam Santoro1* arXiv:2404.02258v1 [cs.LG] 2 Apr 2024 1 Google DeepMind, 2 McGill University & Mila, * Equal Con...
Score: 0.8957221550094074
User feedback: None
Out links: 5167932 Raw text: 5167932https://alignmentforum.org/posts/4rmvMThJYNcCptAya/axrp-episode-22-shard-theory-with-quintin-pope
Title: AXRP Episode 22 - Shard Theory with Quintin Pope — AI Alignment Forum Description: YouTube link • What can we learn about advanced deep learning systems by understanding how humans learn and form values over their lifetimes? Will su… Keywords: No keywords Text content: AXRP Episode 22 - Shard...
Score: 0.8949241482159499
User feedback: None
Out links: 5238754 Raw text: 5238754Title: Reasons compute may not drive AI capabilities growth — AI Alignment Forum Description: How long it will be before humanity is capable of creating general AI is an important factor in discussions of the importance of doing AI alignment r… Keywords: No keywords Text content: Reasons compute may...
Score: 0.8802603803835644
User feedback: None
Out links: 283102 Raw text: 283102https://arxiv.org/pdf/2406.16838
Published in Transactions on Machine Learning Research (11/2024) From Decoding to Meta-Generation: Inference-time Algorithms for Large Language Models Sean Welleck [email protected] Carnegie Mellon University Amanda Bertsch∗ [email protected] arXiv:2406.16838v2 [cs.CL] 20 Nov 2024 Carnegie M...
Score: 0.8768526345536737
User feedback: None
Out links: 5217334 Raw text: 5217334https://alignmentforum.org/posts/W6wBmQheDiFmfJqZy/brain-inspired-agi-and-the-lifetime-anchor
Title: Brain-inspired AGI and the "lifetime anchor" — AI Alignment Forum Description: Last year Ajeya Cotra published a draft report on AI timelines. (See also: summary and commentary by Holden Karnofsky, podcast interview with Ajeya.)… Keywords: No keywords Text content: Brain-inspired AGI and the ...
Score: 0.8680241385120826
User feedback: None
Out links: 5146773 Raw text: 5146773Title: Views on when AGI comes and on strategy to reduce existential risk — AI Alignment Forum Description: Summary: AGI isn't super likely to come super soon. People should be working on stuff that saves humanity in worlds where AGI comes in 20 or 50 years… Keywords: No keywords Text content: View...
Score: 0.8646130516464098
User feedback: None
Out links: 5167935 Raw text: 5167935Title: AXRP Episode 13 - First Principles of AGI Safety with Richard Ngo — AI Alignment Forum Description: YouTube link • This podcast is called AXRP, pronounced axe-urp and short for the AI X-risk Research Podcast. Here, I (Daniel Filan) have conversation… Keywords: No keywords Text content: AXRP E...
Score: 0.8614737847612923
User feedback: None
Out links: 5214739 Raw text: 5214739Title: Birds, Brains, Planes, and AI: Against Appeals to the Complexity/Mysteriousness/Efficiency of the Brain — AI Alignment Forum Description: I argue that an entire class of common arguments against short timelines is bogus, and provide weak evidence that anchoring to the human-brain-human-… Keyw...
Score: 0.8592551080189309
User feedback: None
Out links: 5049361 Raw text: 5049361Title: Why almost every RL agent does learned optimization — AI Alignment Forum Description: Or "Why RL≈RL2 (And why that matters)" … Keywords: No keywords Text content: Why almost every RL agent does learned optimization — AI Alignment Forum This website requires javascript to properly function...
Score: 0.8542357935254249
User feedback: None
Out links: 4909066 Raw text: 4909066https://www.alignmentforum.org/users/adam-jermyn/replies
Title: User Comment Replies — AI Alignment Forum Description: A community blog devoted to technical AI alignment research Keywords: No keywords Text content: User Comment Replies — AI Alignment Forum This website requires javascript to properly function. Consider activating javascript to get acces...
Score: 0.852643768121922
User feedback: None
Out links: 277192 Raw text: 277192https://arxiv.org/pdf/1511.06295.pdf
Under review as a conference paper at ICLR 2016 P OLICY D ISTILLATION arXiv:1511.06295v2 [cs.LG] 7 Jan 2016 Andrei A. Rusu, Sergio Gómez Colmenarejo, Çağlar Gülçehre∗, Guillaume Desjardins, James Kirkpatrick, Razvan Pascanu, Volodymyr Mnih, Koray Kavukcuoglu & Raia Hadsell Google DeepMind Lo...
Score: 0.8477027680946693
User feedback: None
Out links: 267898 Raw text: 267898https://arxiv.org/pdf/1703.03864.pdf
Evolution Strategies as a Scalable Alternative to Reinforcement Learning arXiv:1703.03864v2 [stat.ML] 7 Sep 2017 Tim Salimans Jonathan Ho Xi Chen OpenAI Szymon Sidor Ilya Sutskever Abstract We explore the use of Evolution Strategies (ES), a class of black box optimization algorithms, as an al...
Score: 0.8464387496801347
User feedback: None
Out links: 5167912 Raw text: 5167912Title: AXRP Episode 19 - Mechanistic Interpretability with Neel Nanda — AI Alignment Forum Description: YouTube link • How good are we at understanding the internal computation of advanced machine learning models, and do we have a hope at getting better… Keywords: No keywords Text content: AXRP Epis...
Score: 0.8422426509477547
User feedback: None
Out links: 5167927 Raw text: 5167927Title: AXRP Episode 36 - Adam Shai and Paul Riechers on Computational Mechanics — AI Alignment Forum Description: YouTube link • Sometimes, people talk about transformers as having “world models” as a result of being trained to predict text data on the internet.… Keywords: No keywords Text content: ...
Score: 0.8350078565512994
User feedback: None
Out links: 977534 Raw text: 977534https://www.lesswrong.com/posts/FRv7ryoqtvSuqBxuT/understanding-deep-double-descent
Title: Understanding “Deep Double Descent” — LessWrong Description: Double descent is a puzzling phenomenon in machine learning where increasing model size/training time/data can initially hurt performance, but then i… Keywords: No keywords Text content: Understanding “Deep Double Descent” — LessWro...
Score: 0.830024583613999
User feedback: None
Out links: 1022469 Raw text: 1022469https://www.lesswrong.com/posts/2sLwt2cSAag74nsdN/speaking-to-congressional-staffers-about-ai-risk
Title: Speaking to Congressional staffers about AI risk — LessWrong Description: Orpheus16 shares his experience talking with ~60 congressional staffers about AI risk in May - June 2023. He found staffers were surprisingly open-mi… Keywords: No keywords Text content: Speaking to Congressional staffe...
Score: 0.8294617242126421
User feedback: None
Out links: 1003027 Raw text: 1003027https://www.lesswrong.com/users/rogerdearnaley/replies
Title: User Comment Replies — LessWrong Description: A community blog devoted to refining the art of rationality Keywords: No keywords Text content: User Comment Replies — LessWrong This website requires javascript to properly function. Consider activating javascript to get access to all site func...
Score: 0.8287058255763958
User feedback: None
Out links: 5092358 Raw text: 5092358Title: Full Transcript: Eliezer Yudkowsky on the Bankless podcast — AI Alignment Forum Description: This podcast has gotten a lot of traction, so we're posting a full transcript of it, lightly edited with ads removed, for those who prefer reading ov… Keywords: No keywords Text content: Full Transcri...
Score: 0.823370986383254
User feedback: None
Out links: 274736 Raw text: 274736https://arxiv.org/pdf/2102.06701.pdf
Explaining Neural Scaling Laws Yasaman Bahri∗1 , Ethan Dyer*1 , Jared Kaplan*2 , Jaehoon Lee*1 , and Utkarsh Sharma*†2 1 arXiv:2102.06701v2 [cs.LG] 29 Apr 2024 2 Google DeepMind, Mountain View, CA Department of Physics and Astronomy, Johns Hopkins University [email protected], edyer@google....
Score: 0.8205953582518711
User feedback: None
Out links: 5167920 Raw text: 5167920https://alignmentforum.org/posts/8rBk6fMgwfG4wHt37/axrp-episode-30-ai-security-with-jeffrey-ladish
Title: AXRP Episode 30 - AI Security with Jeffrey Ladish — AI Alignment Forum Description: YouTube link • Top labs use various forms of “safety training” on models before their release to make sure they don’t do nasty stuff - but how robust… Keywords: No keywords Text content: AXRP Episode 30 - AI S...
Score: 0.8203636528101528
User feedback: None
Out links: 977807 Raw text: 977807https://www.lesswrong.com/posts/fRwdkop6tyhi3d22L/there-s-no-such-thing-as-a-tree-phylogenetically
Title: There’s no such thing as a tree (phylogenetically) — LessWrong Description: Trees are not a biologically consistent category. They're just something that keeps happening in lots of different groups of plants. This is a fun fa… Keywords: No keywords Text content: There’s no such thing as a tre...