Yearly

Weekly

Daily

Reading list

Links 25

Title: Birds, Brains, Planes, and AI: Against Appeals to the Complexity/Mysteriousness/Efficiency of the Brain — AI Alignment Forum

Score: 0.8616537922290168

User feedback: None

Out links: 993625 Raw text: 993625

https://www.alignmentforum.org/posts/HhWhaSzQr6xmBki8F/birds-planes-brains-and-ai-against-appeals-to-the-complexity

Title: 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...

Title: Building brain-inspired AGI is infinitely easier than understanding the brain — AI Alignment Forum

Score: 0.8491108986515726

User feedback: None

Out links: 993637 Raw text: 993637

https://www.alignmentforum.org/posts/PTkd8nazvH9HQpwP8/building-brain-inspired-agi-is-infinitely-easier-than

Title: Building brain-inspired AGI is infinitely easier than understanding the brain — AI Alignment Forum Description: Part 1: Brains probably do some useful things in utterly inscrutable ways I'm not so much interested in arguing the strong claim that the brain does… Keywords: No keywords Text cont...

Title: Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training — AI Alignment Forum

Score: 0.8306708795817134

User feedback: None

Out links: 1023430 Raw text: 1023430

https://www.alignmentforum.org/posts/ZAsJv7xijKTfZkMtr/sleeper-agents-training-deceptive-llms-that-persist-through

Title: Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training — AI Alignment Forum Description: I'm not going to add a bunch of commentary here on top of what we've already put out, since we've put a lot of effort into the paper itself, and I'd… Keywords: No keywords Text conte...

Title: None

Score: 0.8270818663345989

User feedback: None

Out links: 2124217 Raw text: 2124217

http://www.cs.toronto.edu/~hinton/absps/googlerectified.pdf

ON RECTIFIED LINEAR UNITS FOR SPEECH PROCESSING M.D. Zeiler1∗ , M. Ranzato2 , R. Monga2 , M. Mao2 , K. Yang2 , Q.V. Le2 , P. Nguyen2 , A. Senior2 , V. Vanhoucke2 , J. Dean2 , G.E. Hinton3 1 New York University, USA 2 Google Inc., USA ABSTRACT Deep neural networks have recently become the gold st...

Title: None

Score: 0.8135172133742868

User feedback: None

Out links: 53248 Raw text: 53248

https://arxiv.org/pdf/2402.00838

OLMo : Accelerating the Science of Language Models α Dirk Groeneveld α α Pete Walsh α Iz Beltagy α α Akshita Bhagia Rodney Kinney Oyvind Tafjord α αβ α Ananya Harsh Jha Hamish Ivison α Ian Magnusson α αβ Yizhong Wang α α Shane Arora David Atkinson Russell Authur Khyathi Ragha...

Title: None

Score: 0.8073852083678074

User feedback: None

Out links: 2124224 Raw text: 2124224

http://www.cs.toronto.edu/~hinton/absps/uai_crbms.pdf

Conditional Restricted Boltzmann Machines for Structured Output Prediction Volodymyr Mnih Department of Computer Science University of Toronto Toronto, Canada Hugo Larochelle ∗ Département d’informatique Université de Sherbrooke Sherbrooke, Canada Abstract Conditional Restricted Boltzmann Machi...

Title: five minute university | Locklin on science | Page 3

Score: 0.8071672731983579

User feedback: None

Out links: 5002640 Raw text: 5002640

https://scottlocklin.wordpress.com/category/five-minute-university/page/3/

Title: five minute university | Locklin on science | Page 3 Description: Posts about five minute university written by Scott Locklin Keywords: No keywords Text content: five minute university | Locklin on science | Page 3 Skip to content Skip to ...

Title: Open problems in Robotics | Locklin on science

Score: 0.7956073191521721

User feedback: None

Out links: 4991805 Raw text: 4991805

https://scottlocklin.wordpress.com/2020/07/29/open-problems-in-robotics/

Title: Open problems in Robotics | Locklin on science Description: Robotics is one of those things the business funny papers regularly wonder about; it seems like consumer robotics is a revolutionary trillion dollar market which is perpetually  20-years away -more or less like nuclear fusion. I had ...

Title: Towards a new kind of science and technology | Locklin on science

Score: 0.7917645097391195

User feedback: None

Out links: 4979095 Raw text: 4979095

https://scottlocklin.wordpress.com/2024/12/18/towards-a-new-kind-of-science-and-technology/

Title: Towards a new kind of science and technology | Locklin on science Description: Reading the history of science and technology, one is struck by how important two branches of physics are: electricity and magnetism, and thermodynamics. The former is mostly how we conveniently pipe energy around....

Title: None

Score: 0.7902377174914134

User feedback: None

Out links: 2124194 Raw text: 2124194

http://www.cs.toronto.edu/~hinton/absps/dropout.pdf

Improving neural networks by preventing co-adaptation of feature detectors G. E. Hinton∗ , N. Srivastava, A. Krizhevsky, I. Sutskever and R. R. Salakhutdinov Department of Computer Science, University of Toronto, 6 King’s College Rd, Toronto, Ontario M5S 3G4, Canada ∗ To whom correspondence should ...

Title: Myths of technological progress | Locklin on science

Score: 0.784407939992258

User feedback: None

Out links: 4987667 Raw text: 4987667

https://scottlocklin.wordpress.com/2009/09/01/myths-of-technological-progress/

Title: Myths of technological progress | Locklin on science Description: I've been doing some writing for Taki's magazine. Most of this writing will be irrelevant or annoying to people who are interested in Science, Technology and Finance. Taki is an old school conservative who is mostly interested ...

Title: None

Score: 0.7784315761978045

User feedback: None

Out links: 2124221 Raw text: 2124221

http://www.cs.toronto.edu/~hinton/absps/multiframe.pdf

Autoregressive product of multi-frame predictions can improve the accuracy of hybrid models Navdeep Jaitly1 , Vincent Vanhoucke2 , Geoffrey Hinton1,2 1 University of Toronto 2 Google Inc. [email protected], [email protected] Abstract We describe a simple but effective way of using multi...

Title: The Bundle Of Holding Is Full of TEETH / That's Some Nice Handiwork • Buttondown

Score: 0.7777022650521453

User feedback: None

Out links: 5265480 Raw text: 5265480

https://buttondown.com/TeethRPG/archive/the-bundle-of-holding-is-full-of-teeth-thats-some/

Title: The Bundle Of Holding Is Full of TEETH / That's Some Nice Handiwork • Buttondown Description: Welcome, News-Riders of The Great Wastes, to the TEETH newsletter! This is a (mostly) weekly transmission about our explorations in the very secret land of... Keywords: No keywords Text content: The ...

Title: Metaphilosophical Mysteries — LessWrong

Score: 0.7766022708345447

User feedback: None

Out links: 206051 Raw text: 206051

https://www.lesswrong.com/posts/MAhueZtNz5SnDPhsy/metaphilosophical-mysteries

Title: Metaphilosophical Mysteries — LessWrong Description: Creating Friendly AI seems to require us humans to either solve most of the outstanding problems in philosophy, or to solve meta-philosophy (i.e., wh… Keywords: No keywords Text content: Metaphilosophical Mysteries — LessWrong This websit...

Title: doi:10.1016/j.tics.2007.09.004

Score: 0.7759493070114146

User feedback: None

Out links: 2124151 Raw text: 2124151

http://www.cs.toronto.edu/~hinton/absps/tics.pdf

Review TRENDS in Cognitive Sciences Vol.11 No.10 Learning multiple layers of representation Geoffrey E. Hinton Department of Computer Science, University of Toronto, 10 King’s College Road, Toronto, M5S 3G4, Canada To achieve its impressive performance in tasks such as speech perception or objec...

Title: How to Think About Activation Patching — AI Alignment Forum

Score: 0.7728593125853405

User feedback: None

Out links: 989135 Raw text: 989135

https://www.alignmentforum.org/posts/xh85KbTFhbCz7taD4/how-to-think-about-activation-patching

Title: How to Think About Activation Patching — AI Alignment Forum Description: This is an excerpt from my post on attribution patching, that I think is of more general interest, around how to think about the technique of activat… Keywords: No keywords Text content: How to Think About Activation Pat...

Title: finance | Locklin on science

Score: 0.7726996108118485

User feedback: None

Out links: 5095935 Raw text: 5095935

https://scottlocklin.wordpress.com/category/finance/

Title: finance | Locklin on science Description: Things dealing with finance Keywords: No keywords Text content: finance | Locklin on science Skip to content Skip to search - Accesskey = s Locklin on science Outsourcing your thoughts Pos...

Title: A Walkthrough of Interpretability in the Wild (w/ authors Kevin Wang, Arthur Conmy & Alexandre Variengien) — AI Alignment Forum

Score: 0.7714470883557178

User feedback: None

Out links: 989150 Raw text: 989150

https://www.alignmentforum.org/posts/DZk6mRo9vhCXN9Rfn/a-walkthrough-of-interpretability-in-the-wild-w-authors

Title: A Walkthrough of Interpretability in the Wild (w/ authors Kevin Wang, Arthur Conmy & Alexandre Variengien) — AI Alignment Forum Description: New paper walkthrough: Interpretability in the Wild: A Circuit for Indirect Object Identification In GPT-2 Small is a really exciting new mechanistic… K...

Title: Leverage numerai signal solution in real trading - Signals - Numerai Forum

Score: 0.7696132420975371

User feedback: None

Out links: 5625857 Raw text: 5625857

https://forum.numer.ai/t/leverage-numerai-signal-solution-in-real-trading/3769

Title: Leverage numerai signal solution in real trading - Signals - Numerai Forum Description: For those, getting positive CORR in weekly signal submission, did any try it in live trading environment t? For example, you get 30 stocks in the portfolio, and based on the regression result from your mod...

Title: None

Score: 0.769124031716929

User feedback: None

Out links: 2124169 Raw text: 2124169

http://www.cs.toronto.edu/~hinton/absps/Outrageously.pdf

Published as a conference paper at ICLR 2017 O UTRAGEOUSLY L ARGE N EURAL N ETWORKS : T HE S PARSELY-G ATED M IXTURE - OF -E XPERTS L AYER Noam Shazeer1 , Azalia Mirhoseini∗†1 , Krzysztof Maziarz∗2 , Andy Davis1 , Quoc Le1 , Geoffrey Hinton1 and Jeff Dean1 1 Google Brain, {noam,azalia,andydavis,qv...

Title: LLM Modularity: The Separability of Capabilities in Large Language Models — AI Alignment Forum

Score: 0.7689446961349281

User feedback: None

Out links: 987123 Raw text: 987123

https://www.alignmentforum.org/posts/j84JhErNezMxyK4dH/llm-modularity-the-separability-of-capabilities-in-large

Title: LLM Modularity: The Separability of Capabilities in Large Language Models — AI Alignment Forum Description: Separating out different capabilities. Post format: First, a 30-second TL;DR, next a 5-minute summary, and finally the full ~40-minute full length te… Keywords: No keywords Text content...

Title: None

Score: 0.7672189088172875

User feedback: None

Out links: 2124228 Raw text: 2124228

http://www.cs.toronto.edu/~hinton/absps/DNN-2012-proof.pdf

Geoffrey Hinton, Li Deng, Dong Yu, George Dahl, Abdel-rahman Mohamed, [Navdeep Jaitly, Andrew Senior, Vincent Vanhoucke, Patrick Nguyen, Tara Sainath, and Brian Kingsbury] IE E Pr E oo f Deep Neural Networks for Acoustic Modeling in Speech Recognition [Four research groups share their views] <AU:...

Title: EIP-3855: PUSH0 instruction - EIPs core - Fellowship of Ethereum Magicians

Score: 0.7657880518074316

User feedback: None

Out links: 1559251 Raw text: 1559251

https://ethereum-magicians.org/t/eip-3855-push0-instruction/7014

Title: EIP-3855: PUSH0 instruction - EIPs core - Fellowship of Ethereum Magicians Description: This is the discussion topic for Keywords: No keywords Text content: EIP-3855: PUSH0 instruction - EIPs core - Fellowship of Ethereum Magicians ...

Title: None

Score: 0.7652138330119215

User feedback: None

Out links: 2124186 Raw text: 2124186

http://www.cs.toronto.edu/~hinton/absps/mcimage.pdf

Generating more realistic images using gated MRF’s Marc’Aurelio Ranzato Volodymyr Mnih Geoffrey E. Hinton Department of Computer Science University of Toronto {ranzato,vmnih,hinton}@cs.toronto.edu Abstract Probabilistic models of natural images are usually evaluated by measuring performance on rat...

Title: The problem with Lee Smolin and physics | Locklin on science

Score: 0.7612267431228579

User feedback: None

Out links: 4962690 Raw text: 4962690

https://scottlocklin.wordpress.com/2009/10/15/the-problem-with-lee-smolin-and-physics/

Title: The problem with Lee Smolin and physics | Locklin on science Description: Lee Smolin wrote a very important book a few years ago called, "the trouble with physics." I give Lee a lot of credit: he recognizes there is trouble with physics; thousands of men have been working on string theory sin...