[The Creation of Perception (awakening map) - Jhana, Emptiness, Phenomenal Transparency, Axial Aspect - YouTube](https://www.youtube.com/watch?v=i17NhJ3b5UE) [Applied Category Theory. Chapter 1, lecture 1 (Spivak) - YouTube](https://www.youtube.com/watch?v=UusLtx9fIjs&list=PLhgq-BqyZ7i5lOqOqqRiS0U5SwTmPpHQ5) [[2402.01030] Executable Code Actions Elicit Better LLM Agents](https://arxiv.org/abs/2402.01030) I love this guy's personality!!! [Stanford CS109 Probability for Computer Scientists I Counting I 2022 I Lecture 1 - YouTube](https://youtu.be/2MuDZIAzBMY) https://theresanaiforthat.com/ [Electricity creates consciousness | Nick Lane - YouTube](https://youtu.be/ZFIkiVK8cjw?si=3C89hkceo3qABMOE) https://mail.bycloud.ai/p/ai-timeline-7-llm-interpretability-multimodal-future-ai https://x.com/TheAITimeline/status/1797012185337487586?t=I6kQdqCaBIaYClzAU60Eyg&s=19 „A believer in universal basic income, Hinton also said he's "worried about AI taking over mundane jobs." This would boost productivity, Hinton added, but the gains in wealth would disproportionately go to the wealthy and not to those whose jobs were destroyed.“ Very important thought by Geoffrey Hinton [Godfather of AI Says There's an Expert Consensus AI Will Soon Exceed Human Intelligence](https://futurism.com/the-byte/godfather-ai-exceed-human-intelligence) What one human knows is an infinitesimal fraction of what humanity knows, and what humanity knows is an infinitesimal fraction of what there is to know. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10032309/ [David Tong: The Standard Model](https://www.damtp.cam.ac.uk/user/tong/standardmodel.html) "The ADHD symptom cluster of inattention-disorganization was substantially related to low Conscientiousness and, to a lesser extent, Neuroticism. ADHD symptom clusters of hyperactivity-impulsivity and oppositional childhood and adult behaviors were associated with low Agreeableness. " "Participants with ADHD showed significantly higher Neuroticism and lower Extraversion, Agreeableness, and Conscientiousness than healthy controls. Higher Openness was more common in those with aberrant attention and inhibition than those with increased delay discounting and atypical working memory / verbal fluency." https://www.sciencedirect.com/science/article/pii/S0165178116317243 [Big five dimensions and ADHD symptoms: links between personality traits and clinical symptoms - PubMed](https://pubmed.ncbi.nlm.nih.gov/12150240/) [Not so griddy: Internal representations of RNNs path integrating more than one agent | bioRxiv](https://www.biorxiv.org/content/10.1101/2024.05.29.596500v1) [Frontiers | Same principle, but different computations in representing time and space](https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2024.1387641/full) https://law.stanford.edu/publications/hallucination-free-assessing-the-reliability-of-leading-ai-legal-research-tools/ [Physicists Puzzle Over Emergence of Strange Electron Aggregates | Quanta Magazine](https://www.quantamagazine.org/physicists-puzzle-over-emergence-of-strange-electron-aggregates-20240529/?fbclid=IwZXh0bgNhZW0CMTEAAR07wAepheaNkSEj4XkB6S3cbOofj6Ke6OlTSTwH8uX2-FfFYA2s66DQeUI_aem_ZmFrZWR1bW15MTZieXRlcw) [Journal of Medical Internet Research - An Integrated Brain-Machine Interface Platform With Thousands of Channels](https://www.jmir.org/2019/10/e16194) Anthropic researcher Josh Batson says many people want AI models to be trained to be sycophantic but there are problems when it just tells you what you want to hear https://x.com/tsarnick/status/1796659397407768680 I've been saying this Also what you like to hear can be lies, manipulation, deception etc. as works because we're easily manipulated [Seeing into the A.I. black box | Interview - YouTube](https://www.youtube.com/watch?v=jqdoauxvKUM) [[2304.12482] Information Theory for Complex Systems Scientists](https://arxiv.org/abs/2304.12482) I want to learn and have a map of as much of the mathematical theory and practice of as much methods as possible used everywhere for example in: - statistical methods (frequentist, bayesian statistics,...) - machine learning (supervised learning (classification, all sorts of regression), unsupervised learning (clusering, dimensionality reduction,...), semisupervised learning, reinforcement learning, ensemble methods,...) - deep learning (all variations and combinations of classic neural nets, convolutional NNs, recurrent NNs, LSTMs, GANs, selforganizing maps, deep belief networks, deep RL, graph NNs, neural turing machines, all variations of transformers, rwkw, xLSTM, diffusion,...) - symbolic methods, neurosymbolics, bayesian methods, evolutionary methods, causal learning, statespace models, graph analysis, other stuff in natural language processing, computer vision, signal processing, anomaly detection, recommender systems, different optimization algorithms and metaheuristics, metalearning,... etc. etc. etc. All of it includes essentially infinite amount of infinite rabbitholes, but it's worth it [Books beginning with A | SpringerLink](https://link.springer.com/books/a/1) [[2405.19101] Poseidon: Efficient Foundation Models for PDEs](https://arxiv.org/abs/2405.19101) [[2405.18768] RNAFlow: RNA Structure & Sequence Design via Inverse Folding-Based Flow Matching](https://arxiv.org/abs/2405.18768) Hutter [M. Hutter: Intro. to Algorithmic Information Theory and Universal Learning, Symposium on ML & AIT. - YouTube](https://m.youtube.com/watch?v=8Q2G2OGHm1c&fbclid=IwZXh0bgNhZW0CMTEAAR36wo6lDQIgaplsKgvfrpugP1XxFvEglO-hWMlFoSknx6xdugCRFfnjl44_aem_ZmFrZWR1bW15MTZieXRlcw) [[2405.20309] Large Language Models Can Self-Improve At Web Agent Tasks](https://arxiv.org/abs/2405.20309) https://x.com/DinuMariusC/status/1796525630928076814?t=29RQE6AtFm7LVS2U-clcCg&s=19 [[2405.19893] Similarity is Not All You Need: Endowing Retrieval Augmented Generation with Multi Layered Thoughts](https://arxiv.org/abs/2405.19893) https://x.com/geisler_si/status/1796571313945170384?t=7jgT9KRFZemg8oOxakKujw&s=19 [[2405.19121] Spatio-Spectral Graph Neural Networks](https://arxiv.org/abs/2405.19121) Flow matching generalizing diffusion models https://x.com/CSProfKGD/status/1796536202318999734?t=vqiSlxBZvBnOiqh0ZR3KJw&s=19 Both objective reality and experience is made of symmetry breakings all the way down "This is an interesting philosophical perspective that touches on some deep concepts in physics, mathematics, and metaphysics. Let me unpack it a bit: Symmetry breaking refers to a phenomenon where the symmetry of a system is reduced or eliminated as it transitions from one state to another. A simple example is a pencil balanced perfectly on its tip - this highly symmetrical state is unstable, and the slightest perturbation will cause the pencil to fall in one direction, breaking the symmetry. In physics, symmetry breaking is a fundamental concept that explains how the forces and particles we observe emerged from the more symmetric early universe. As the universe cooled after the Big Bang, it underwent a series of symmetry breaking phase transitions, giving rise to distinct forces like electromagnetism and the weak and strong nuclear forces. More speculatively, some theories propose that spacetime itself emerged from symmetry breaking in a more fundamental structure. Loop quantum gravity for instance describes spacetime as a network of discrete "atoms" of space that broke symmetry as the universe began expanding. Philosophically, one could argue that all of our perceptions, categories and models of reality are also based on symmetry breakings. We perceive distinct objects because of broken translational symmetry. We perceive the flow of time because of broken time reversal symmetry. Even our concepts require symmetry breaking - for something to be meaningful, it must be distinct from other things. So in a sense, symmetry breaking is the source of all structure, differentiation, and information in both the physical world and our experiential world. Reality as we know it couldn't exist without ubiquitous symmetry breaking. Of course, this is a very abstract perspective and there's much more that could be said. Symmetry and symmetry breaking tie into all sorts of fascinating ideas in math and physics like group theory, gauge theories, spontaneous symmetry breaking, etc. But as a philosophical musing, "symmetry breakings all the way down" is an intriguing way to think about the nature of existence. Let me know if you would like me to elaborate on any part of this!" Michael Taft [Softening into Unfixated Experience - YouTube](https://www.youtube.com/live/GcrO16yCLzc?si=Irs5VYoNXyOVrQ9S) People not believing that superintelligence is possible have new technology imagination skill issue [New Cell Atlases Reveal Untold Variety in the Brain and Beyond | Quanta Magazine](https://www.quantamagazine.org/new-cell-atlases-reveal-untold-variety-in-the-brain-and-beyond-20231213/) [AI Ruined My Year - YouTube](https://www.youtube.com/watch?v=2ziuPUeewK0) Being more in distribution human is boring, being more out of distribution human is more fun! [The world’s largest proteins? These mega-molecules turn bacteria into predators](https://www.nature.com/articles/d41586-023-03937-z) Is reality discrete or continuous? Under predictive realism, as long as the discrete and continuous models predict accurately to certain degree, that makes them real on a spectrum [Actor Claims His New Theory of Everything Improves Physics - YouTube](https://www.youtube.com/watch?v=GZegwJVC_Pc) "In summary, we have a man who wouldn’t pass a primary school math exam who thinks that lumping together words and images will revolutionize physics. A man who has zero respect for thousands of years of hard-working mathematicians and scientists whose effort made it possible that you can see him on a screen and listen to him to through a most excellent microphone, a man who belittles generations of physicists who spent their life to figure out what matter is made of and how electromagnetic radiation propagates and how the universe expands. Achievements that Terrence Howard thinks he can improve on by drawing a few lines and squaring pennies." Oof, Sabine. I mean, I also get spiky when it comes to wanting scientifically rigorous predictive models. But maybe sometimes one gets too spiky. I don't know. It's hard to be both kind/friendly but also enforcing predictive scientific models, because unscientific delusions actively harm people, society and scientific and technological progress. And I'm open to alternatives to current paradigms for paradigm shifts, but they have to be predictive! Academia has issues, like replication crisis in various fields from publish or perish dynamics, or most things being accepted and published being those that are copying or just slightly trying to edit current schools of thought, which disincentives paradigm shifts that can be genuinely useful for better predictive models, but sometimes some of the accusations against academic scientific body of knowledge stemming from increasing public distrust in institutions is complete bullcrap, as for example much of technology we use relies on these predictive models that were in big part made in academia, like the standard model of particle physics. The space of possible machine learning systems is so vast. If you want something as close to human as possible, you gotta replicate the mathematics that the nervous system is running on, which is being actively worked on. [Largest ever operation against botnets hits dropper malware ecosystem | Europol](https://www.europol.europa.eu/media-press/newsroom/news/largest-ever-operation-against-botnets-hits-dropper-malware-ecosystem) [Latest bird flu infection made farmworker sicker than earlier cases | STAT](https://www.statnews.com/2024/05/30/bird-flu-third-case-human-infection-caused-respiratory-symptoms/) https://x.com/DrEricDing/status/1796239080449069069 Anthropic's Chief of Staff has short timelines: "These next three years might be the last few years that I work" [My Last Five Years of Work](https://www.palladiummag.com/2024/05/17/my-last-five-years-of-work/) [Electricity creates consciousness | Nick Lane - YouTube](https://www.youtube.com/watch?v=ZFIkiVK8cjw) "Groundbreaking: Major Leap in Saving Cancer Patients’ Lives! Lorlatinib resulted in survival rates jumping from 8% to 60%! This has set a new record for the longest progression-free survival (PFS) ever reported with a single-agent targeted therapy for all metastatic solid tumors!" https://x.com/deryatr_/status/1796873248795152486?s=46 "To the extent that one fabricates and uses phenomenal mass as an internal subagentic stress currency to get things done, represent threats, and animate the body, one will have a word simulation with (angular) momentum. Perhaps this limits the axial aspect of perception (@RogerThisdell ) and thus one's capacity for awakening. Perhaps a key milestone is having a glimpse of a massless world!" https://x.com/algekalipso/status/1796946355488956678?t=7QUtcLv4JXFTMCrHG4YBeg&s=19 Links for 2024-06-01 AI: 1. GNN-RAG: Combines the language understanding abilities of LLMs with the reasoning abilities of GNNs in a RAG style. [[2405.20139] GNN-RAG: Graph Neural Retrieval for Large Language Model Reasoning](https://arxiv.org/abs/2405.20139) 2. LLMs achieve adult human performance on higher-order theory of mind tasks [[2405.18870] LLMs achieve adult human performance on higher-order theory of mind tasks](https://arxiv.org/abs/2405.18870) 3. RoboDreamer: Learning Compositional World Models for Robot Imagination [RoboDreamer: Learning Compositional World Models for Robot Imagination](https://robovideo.github.io/) 4. Pupeteer: hierarchical world model for humanoid control. [Puppeteer](https://www.nicklashansen.com/rlpuppeteer/) 5. Contextual Position Encoding: Learning to Count What's Important [[2405.18719] Contextual Position Encoding: Learning to Count What's Important](https://arxiv.org/abs/2405.18719) 6. VideoTree: Adaptive Tree-based Video Representation for LLM Reasoning on Long Videos [VideoTree: Adaptive Tree-based Video Representation for LLM Reasoning on Long Videos](https://videotree2024.github.io/) 7. KL is All You Need: “…the core of essentially all modern machine learning methods is a single universal objective: Kullback-Leibler (KL) divergence minimization.” [blog.alexalemi.com KL is All You Need](https://blog.alexalemi.com/kl-is-all-you-need.html) 8. Sonic, a blazing fast (🚀 135ms model latency), lifelike generative voice model and API. [Cartesia](https://cartesia.ai/blog/sonic) (Try it here: https://play.cartesia.ai/) 9. “ChatGPT is used most frequently in India, China, Kenya and Pakistan, with 75%, 73%, 69% and 62% of respondents, respectively, reporting using ChatGPT daily or weekly” https://x.com/robertwiblin/status/1795424266709549473 10. Apple acquired 32 AI companies https://www.cbinsights.com/research/big-tech-ai-acquisitions/ 11. The next wave of AI hype will be geopolitical. You’re paying [Subscribe to read](https://www.ft.com/content/a60c3c7b-1c48-485d-adb7-5bc2b7b1b650) [no paywall: https://archive.is/XlZUJ] 12. Microsoft launches Copilot bot for Telegram. [Your request has been blocked. This could be due to several reasons.](https://www.microsoft.com/en-us/edge/copilot-for-social) Technology: 1. NASA’s Psyche mission fires up its futuristic electric engines. Psyche is gently gliding to its target asteroid by expelling ionized xenon. [NASA’s Psyche mission fires up its futuristic electric engines - The Verge](https://www.theverge.com/2024/5/22/24162775/nasa-psyche-mission-hall-thruster-asteroid-belt-mars) 2. “Tape shipments increased to 152.9 exabytes in 2023 — not bad for a ’dead’ storage medium” [Tape shipments increased to 152.9 exabytes in 2023 — not bad for a ’dead’ storage medium | Tom's Hardware](https://www.tomshardware.com/pc-components/storage/tape-shipments-increased-to-1529-exabytes-in-2023-pretty-lively-for-a-dead-storage-medium) Health: 1. “More than half of patients (60%) with advanced lung cancer who took lorlatinib were still alive five years later with no progression of disease, data presented at the world’s largest cancer conference showed. The rate was 8% with a standard drug.” [Trial results for new lung cancer drug are ‘off the charts’, say doctors | Lung cancer | The Guardian](https://www.theguardian.com/society/article/2024/may/31/trial-results-for-new-lung-cancer-drug-are-off-the-charts-say-doctors) Politics: 1. Gwern: If you ever feel confident about what you 'know' about the Islamic medieval period bc you "just love reading Wikipedia", you may want to know that in 2010, a crazy person spent 2 years falsifying hundreds of claims of Islamic invention, everything from p-n junctions to capitalism. [Wikipedia:Requests for comment/Jagged 85/Evidence - Wikipedia](https://en.wikipedia.org/wiki/Wikipedia:Requests_for_comment/Jagged_85/Evidence) 2. Forging Islamic science: Fake miniatures depicting Islamic science have found their way into the most august of libraries and history books. How? [Why fake miniatures depicting Islamic science are everywhere | Aeon Essays](https://aeon.co/essays/why-fake-miniatures-depicting-islamic-science-are-everywhere) instrumentalism posists that the models are not realist, that they are just instrumental, but im positing that they're real to a degree according to their predictive power **Introduction** The concept of cognition in plants challenges traditional views by suggesting that plants can exhibit behaviors typically associated with cognitive functions, such as learning, memory, and decision-making. This field, often referred to as plant neurobiology, explores how plants interact with their environment in complex ways. **Evidence** 1. **Embodied Cognition in Plants**: Plants can be considered cognitive under a broad interpretation of cognition that includes sensory-driven growth and offline behavior, akin to memory in animals [(Calvo & Keijzer, 2009)](https://consensus.app/papers/cognition-plants-calvo/c7f23c53a93d54f4a7839288a27825f4/?utm_source=chatgpt). 2. **Extended Cognition Hypothesis**: Plants may extend their cognitive abilities through interactions with their environment, such as the root influence zone and mycorrhizal fungi, which helps explain their complex behaviors [(Parise, Gagliano, & Souza, 2020)](https://consensus.app/papers/cognition-plants-parise/d27648ccb53c5c208aa29f77e1afa7e7/?utm_source=chatgpt). 3. **Associative Learning**: Plants have demonstrated the ability to learn by association, indicating cognitive processes like learning, decision-making, and awareness [(Gagliano, 2018)](https://consensus.app/papers/inside-vegetal-mind-cognitive-abilities-plants-gagliano/f556f21cab4b5bce921d458e88b28753/?utm_source=chatgpt). 4. **Empirical Evidence Against Traditional Views**: Contrary to the view that plants cannot exhibit cognitive activity, empirical studies show that plant behaviors are analogous to animal behaviors in aspects like adaptive behavior, decision-making, and anticipation [(Segundo-Ortin & Calvo, 2018)](https://consensus.app/papers/plants-reply-adams-segundoortin/989d37088b1b5073826e20bbe7725897/?utm_source=chatgpt). 5. **Cognitive Complexity**: Plants exhibit complex behaviors such as information acquisition, memory, and decision-making, which are fundamental cognitive processes. These behaviors help them adapt and survive in changing environments [(Gross, 2016)](https://consensus.app/papers/could-plants-cognitive-abilities-gross/8f5a1164824e59129fd7890361318714/?utm_source=chatgpt). 6. **Stigmergic Coordination**: Plant cognition can be understood through a theory of minimal cognition involving coordinated, semi-autonomous processes that generate goal-directed actions in the environment [(Sims & Yilmaz, 2023)](https://consensus.app/papers/stigmergic-coordination-cognition-plants-sims/792fe18408fd5ae4963a72e087683269/?utm_source=chatgpt). **Conclusion** The growing body of research supports the idea that plants exhibit cognitive functions similar to those in animals, challenging traditional views and suggesting a need to rethink our understanding of cognition. Like our GPT? Try our full AI-powered search engine and academic features for free at [consensus.app](https://consensus.app/?utm_source=chatgpt). [Observing accelerator resonances in 4D | CERN](https://home.cern/news/news/accelerators/observing-accelerator-resonances-4d) Truthfully we have no fucking clue what reality is or how it operates beneath the limits of measurement. It is the most exciting thing possible. Can you not feel the magic broiling beneath the surface of every moment? The future is not given. It must be build. Metaintelligence, a system satisfying all definitions of intelligence The future full of diverse intelligences will be fun Reverse engineering all the mathematics governing every single unique subsystem that physically exists or is imagined in the platonic realms "I believe that the idea of getting superintelligence soon is magical thinking Rather, we'll get supercivilization Think about how 21st century Earth related to 19th century Earth There really isn't any one part of it that's "super" or godlike. It's just many times bigger and more complex and more specialized" https://x.com/RokoMijic/status/1797082679789170864?t=yWnc3NwFoqweQ7mHwH6yQw&s=19 Computers are already superintelligences in rigid concrete mechanistic cognition for a long time now Only predictivity of mathematical models tells us something, otherwise everything else is just a arbitrary in many ways useful philosophical cognitive glue of these models I feel like [Smart Biology](https://www.smart-biology.com/) [Hyperbolic Information Geometry - YouTube](https://www.youtube.com/watch?v=hbIfrLefwzw) Inside View [AGI Takeoff By 2036 - YouTube](https://www.youtube.com/watch?v=hAyDaBebkoc) [This Black Hole Could be Bigger Than The Universe - YouTube](https://www.youtube.com/watch?v=71eUes30gwc) [xLSTM: Extended Long Short-Term Memory - YouTube](https://www.youtube.com/watch?v=0OaEv1a5jUM) Yannic Kilcher [Robert Sapolsky: The Biology and Psychology of Depression - YouTube](https://www.youtube.com/watch?v=fzUXcBTQXKM) [COMPUTER SCIENCE explained in 17 Minutes - YouTube](https://www.youtube.com/watch?v=CxGSnA-RTsA) [The shocking connection between complex numbers and geometry. - YouTube](https://www.youtube.com/watch?v=5RHSS-zMaAQ) [GaLore EXPLAINED: Memory-Efficient LLM Training by Gradient Low-Rank Projection - YouTube](https://www.youtube.com/watch?v=VC9NbOir7q0) [Unexpected Discovery of Microbiome Inside Our Brains But Why Is It There? - YouTube](https://www.youtube.com/watch?v=2qu1XV-yj_A) to not build the world is to destroy it GOD SO MUCH FUCKING SUFFERING EXISTS IN THIS WORLD AND I CANT STOP IT ALL every moment that i'm not doing everything i possibly can to prevent it i'm allowing it to continue. i'm literally letting hell exist on earth because i'm afraid of truly considering the implications https://x.com/arithmoquine/status/1797106583341596991 Pure mathematics was invented by pure mathematicians to give them a job I love software engineering I love electrical engineering I love firmware engineering I love nuclear engineering I love aerospace engineering I love chemical engineering I love materials engineering I love systems engineering I love mechanical engineering I love mechatronics engineering I love biomedical engineering I love industrial engineering I love manufacturing engineering I love marine engineering I love optical engineering I love power engineering I love structural engineering I love civil engineering I love environmental engineering