0 Members and 2 Guests are viewing this topic.
Quote from: hamdani yusuf on 09/02/2025 15:13:31It's about cognitive offloading.You can focus on what matters to you the most.Integrity, credibility, traceability, relevance, resilience, brevity. Which, as you have so amply demonstrated, are not characteristics of chatbot output.
It's about cognitive offloading.You can focus on what matters to you the most.
Timestamps: 0:00 - About Terence Tao and the Distance Ladder2:02 - Earth8:07 - Moon11:15 - Sun15:45 - Heliocentrism in Antiquity18:27 - Kepler?s genius27:16 - Where this leaves usTerence kindly put together this FAQ with added details and corrections, which I'll post and update in this pinned comment.* 4:50 How did Eratosthenes know that the Sun was so far away that its light rays were close to parallel?This was not made so clear in our discussions or in the video (other than a brief glimpse of the timeline at 18:27), but Eratosthenes's work actually came after Aristarchus, so it is very likely that Eratosthenes was aware of Aristarchus's conclusions about how distant the Sun was from the Earth. Even if Aristarchus's heliocentric model was disputed by the other Greeks, at least some of his other conclusions appear to have attracted some support. Also, after Eratosthenes's time, there was further work by Greek, Indian, and Islamic astronomers (such as Hipparchus, Ptolemy, Aryabhata, and Al-Battani) to measure the same distances that Aristarchus did, although these subsequent measurements also were somewhat far from modern accepted values.* 5:17 is it completely accurate to say that on the summer solstice, the Earth's axis of rotation is tilted "directly towards the Sun"?Strictly speaking, "in the direction towards the Sun" is more accurate than "directly towards the Sun"; it tilts at about 23.5 degrees towards the Sun, but it is not a total 90-degree tilt towards the Sun.* 7:27 Are the riverboat merchants and the "grad student" the leading theories for how Eratosthenes measured the distance from Alexandria to Syene?There is some recent research that suggests that Eratosthenes may have drawn on the work of professional bematists (step measurers) for this calculation. This somewhat ruins the "grad student" joke, but perhaps should be disclosed for the sake of completeness.* 8:51 How long is a "lunar month" in this context? Is it really 28 days?In this context the correct notion of a lunar month is a "synodic month" - the length of a lunar cycle relative to the Sun - which is actually about 29 days and 12 hours. It differs from the "sidereal month" - the length of a lunar cycle relative to the fixed stars - which is about 27 days and 8 hours - due to the motion of the Earth around the Sun (or the Sun around the Earth, in the geocentric model). [A similar correction needs to be made around 14:59, using the synodic month of 29 days and 12 hours rather than the "English lunar month" of 28 days (4 weeks).]* 10:47 Is the time taken for the Moon to complete an observed rotation around the Earth slightly less than 24 hours as claimed?Actually, I made a sign error: the lunar day (also known as a tidal day) is actually 24 hours and 50 minutes, because the Moon rotates in the same direction as the spinning of Earth around its axis. The animation therefore is also moving in the wrong direction as well (related to this, the line of sight is covering up the Moon in the wrong direction to the Moon rising at around 10:38).* 17:37 Could the parallax problem be solved by assuming that the stars are not distributed in a three-dimensional space, but instead on a celestial sphere?Putting all the stars on a fixed sphere would make the parallax effects less visible, as the stars in a given portion of the sky would now all move together at the same apparent velocity - but there would still be visible large-scale distortions in the shape of the constellations because the Earth would be closer to some portions of the celestial sphere than others. (This problem would be solved if the celestial sphere was somehow centered around the moving Earth rather than the fixed Sun, but then this basically becomes the geocentric model with extra steps.)* 18:29 Did nothing of note happen in astronomy between Eratosthenes and Kepler?Not at all! There were significant mathematical, technological, theoretical, and observational advances by astronomers from many cultures (Greek, Islamic, Indian, Chinese, European, and others) during this time, for instance improving some of the previous measurements on the distance ladder, a better understanding of eclipses, axial tilt, and even axial precession, more sophisticated trigonometry, and the development of new astronomical tools such as the astrolabe. But in order to make the overall story of the cosmic distance ladder fit into a two-part video, we chose to focus primarily on the first time each rung of the ladder was climbed.* 19:07 Isn't it tautological to say that the Earth takes one year to perform a full orbit around the Sun?Technically yes, but this is an illustration of the philosophical concept of "referential opacity": the content of a sentence can change when substituting one term for another (e.g., "1 year" and "365 days"), even when both terms refer to the same object. Amusingly, the classic illustration of this, known as Frege's puzzles, also comes from astronomy: it is an informative statement that Hesperus (the evening star) and Phosphorus (the morning star) are the same object (which nowadays we call Venus), but it is a mere tautology that Hesperus and Hesperus are the same object: changing the reference from Phosphorus to Hesperus changes the meaning.* 19:10 How did Copernicus figure out the crucial fact that Mars takes 687 days to go around the Sun? Was it directly drawn from Babylonian data?Technically, Copernicus drew from tables by Islamic astronomers, which were in turn based on earlier tables by Greek astronomers, who also incorporated data from the ancient Babylonians, so it is more accurate to say that Copernicus relied on centuries of data, at least some of which went all the way back to the Babylonians.Among all of this data was the times when Mars was in opposition to the Sun; if one imagines the Earth and Mars as being like runners going around a race track circling the Sun, with Earth on an inner track and Mars on an outer track, oppositions are analogous to when the Earth runner "laps" the Mars runner. From the centuries of observational data, such "laps" were known to occur about once every 780 days (this is known as the synodic period of Mars). Because the Earth takes 365 days to perform a "lap", it is possible to do a little math and conclude that Mars must therefore complete its own "lap" in 687 days (this is known as the sidereal period of Mars).* 23:28 Can one work out the position of Earth from fixed locations of the Sun and Mars when the Sun and Mars are in conjunction (the same location in the sky) or opposition (opposite locations in the sky)?Technically, these are two times when the technique of triangulation fails to be accurate; and also in the former case it is extremely difficult to observe Mars due to the proximity to the Sun. But again, following the Universal Problem Solving Tip from 23:07, one should initially ignore these difficulties to locate a viable method, and correct for these issues later.* 24:21 Did Brahe have exactly 10 years of data on Mars's positions?Actually, it was more like 17 years, but with many gaps, due both to inclement weather, as well as Brahe turning his attention to other astronomical objects than Mars in some years; also, in times of conjunction, Mars might only be visible in the daytime sky instead of the night sky, again complicating measurements. So the "jigsaw puzzle pieces" in 25:26 are in fact more complicated than always just five locations equally spaced in time; there are gaps and also observational errors to grapple with. But to understand the method one should ignore these complications; again, see "Universal Problem Solving Tip #1". Even with his "idea of true genius" (which, incidentally one can find in Einstein's introduction to Carola Baumgardt's "Life of Kepler"), it took many years of further painstaking calculation for Kepler to tease out his laws of planetary motion from Brahe's messy and incomplete observational data.
I think it's important to know how we've got our current model of the universe.
Quote from: hamdani yusuf on 11/02/2025 12:32:54I think it's important to know how we've got our current model of the universe.In a word, without AI.
0:00 How to determine protein structures3:50 Why are proteins so complicated?5:34 The CASP Competition and Deep Mind 9:08 How does Alphafold work?12:06 3 ways to get better AI14:24 What is a Transformer in AI? 17:15 The Structure Module18:35 Alphafold 2 wins the Nobel Prize20:36 Designing New Proteins - RF Diffusion22:58 The Future of AI
The model for protein structure was solved with AI, as shown in this video
Reasoning models are a new category of specialized language models. We now have models that push the state of the art for logical reasoning to find answers.IntroductionReasoning models are a new category of specialized language models. They are designed to break down complex problems into smaller, manageable steps and solve them through explicit logical reasoning (This step is also called ?thinking?). Unlike general-purpose LLMs which might generate direct answers, reasoning models are specifically trained to show their work and follow a more structured thought process. Some models don?t show their logical reasoning phase while others explicitly show their logical reasoning phase. The reasoning phase shows how the model can break down the problem stated into smaller problems (Decomposition), try different approaches (Ideation), choose the best approaches (validation), reject invalid approaches (possibly backtracking) and finally choose the best answer (execution/solving). Reasoning models like OpenAI?s o1, o1-mini, o3-mini, and DeepSeek-R1 are available in the AI Toolkit for VS Code model catalog, Azure AI Foundry model catalog as well as online for free usage (rate-limited) in the Github Marketplace for models.
What will Software Engineering look like at the end of 2025? After lots of research and analysis, I'm breaking down the 4 massive changes that are transforming how we build software. The landscape of software development is evolving faster than ever, with AI agents revolutionizing workflows, real-time data processing pushing the boundaries of what's possible, and legacy systems undergoing the biggest modernization effort in history. Whether you're a seasoned developer, a tech enthusiast, or just curious about where the industry is heading, this deep dive will show you why 2025 is shaping up to be a pivotal year for software engineering. From AI teams that collaborate with developers like never before, to systems that learn and adapt like a close friend, we're about to witness a transformation that will redefine not just how we build software, but what software fundamentally is.Timestamps:00:00 Introduction: The Future of Software Engineering00:07 Sleepless Nights: Fascination with 2025 Trends00:25 Industry Insights: Tech Leaders' Predictions00:39 AI Systems: The New Focus00:52 A Moment at a Major Tech Company01:31 Big Idea #1: Software That Learns Like a Friend02:03 Real-Life Applications of Software 2.002:59 Big Idea #2: The AI Partner Revolution03:45 AI Agents: Specialized Teams for Developers05:05 Balancing AI and Human Creativity06:31 Big Idea #3: The Great Digital Renovation07:53 Modernizing Legacy Systems Safely08:06 Big Idea #4: Speed of Light Revolution08:58 Real-Time Data: Challenges and Security09:30 Conclusion: The Future of Software Engineering
A new open-source AI model, OpenThinker-32B, has outperformed DeepSeek R1 and other major models despite using far fewer resources. It excels in math, coding, and logical reasoning, proving that smarter training methods can beat brute-force approaches. Another breakthrough, Huginn-3.5B, introduces latent reasoning and a unique recurrent depth technique, allowing it to refine answers internally without requiring massive computational power.🔍 Key Topics: The open-source AI model OpenThinker-32B outperforming DeepSeek R1 with a smarter approach How latent reasoning and recurrent depth in Huginn-3.5B are redefining AI problem-solving The shift from brute-force training to efficiency-driven AI models that challenge industry giants 🎥 What?s Inside: How OpenThinker-32B beats proprietary models despite using fewer resources Why Huginn-3.5B?s hidden loops enable deeper reasoning without massive computation The impact of open-source AI innovation on coding, math, and logical reasoning benchmarks 📊 Why It Matters: This video explores groundbreaking advances in *AI reasoning, open-source models, and efficiency-driven training*, revealing how smaller but smarter AI models are reshaping the landscape of artificial intelligence.
The title is misleading. Deepseek still scores higher.Open thinker did better in the Math and one other benchmark. So DeepSeek does not score higher in everything.It's an endless race; no one is a real winner, only in front for now.
To get smarter, traditional AI models rely on exponential increases in the scale of data and computing power. Noam Brown, a leading research scientist at OpenAI, presents a potentially transformative shift in this paradigm. He reveals his work on OpenAI's new o1 model, which focuses on slower, more deliberate reasoning ? much like how humans think ? in order to solve complex problems. (Recorded at TEDAI San Francisco on October 22, 2024)
"The Grok team didn't talk about open source"Not true! Elon said in the livestream that they plan to open source old models once the new models are up and running. Basically the way ID used to work with their game engines.That means we'll be getting open source Grok 2 before too long.
Google?s new AI breakthrough, Mixture-of-Depths (MoD), makes transformer models faster and more efficient by skipping unnecessary computations, focusing only on important words in a sequence. This innovation reduces processing costs while maintaining or even improving AI performance, allowing models to train longer or scale bigger within the same budget. By combining MoD with Mixture-of-Experts (MoE), Google has created an even smarter system that optimizes computing power, leading to faster AI language processing and significant improvements in machine learning efficiency.🔍 Key Topics: How Google?s Mixture-of-Depths (MoD) AI makes language models faster and more efficient The breakthrough method that skips unnecessary computations while maintaining top performance How MoD, combined with Mixture-of-Experts (MoE), optimizes AI processing power 🎥 What?s Inside: Why Google?s new AI revolutionizes language models by focusing only on important words How this breakthrough cuts processing costs, speeds up AI, and improves efficiency The impact of MoD on Google?s AI advancements and the future of deep learning 📊 Why It Matters: This video explores Google?s latest AI innovation, which boosts speed, reduces computational waste, and improves AI efficiency, setting a new standard for transformer-based models and reshaping the future of machine learning.
Terence Tao on how we measure the cosmos | Part 1
How we know the distances to the planets, stars, and faraway galaxies.Timestamps: 0:00 - Intro1:04 - Distance to Venus7:30 - Speed of light10:08 - Nearby stars13:20 - The Milky Way18:14 - Nearby Galaxies19:44 - Distant Galaxies22:42 - Lingering mysteries
Asteroid 2024 YR4, has sparked concern about its chance of hitting Earth in December 2032. How worried should we be?0:00 The Discovery 2024 YR40:56 How to Spot an Asteroid3:07 Ad Read4:33 Why Are We So Bad at Predicting Asteroid Impacts?11:42 How Much Damage Could YR4 Do?13:04 How Could We Stop Asteroid YR4?16:34 Conclusion
Elon's Grok-3 Just Beat EVERYONE?!
Evaluating Grok3: Hype or Real Deal?Is Grok-3 really the best AI model, or did xAI manipulate the benchmarks? In this video, we break down the controversy surrounding Grok-3?s performance claims, comparing it to OpenAI?s O3 Mini and analyzing real-world testing results. Watch as we dive into its reasoning capabilities, explore its "Think" mode, and see if it truly outperforms the competition!Timestamps:00:00 Introduction: Grok3 Controversy00:29 OpenAI's Critique and Grok3's Response01:16 Analyzing Grok3's Performance02:41 Grok3's Reasoning Capabilities03:09 Testing Grok3 on Misguided Attention
Explore two learning algorithms for neural networks: stochastic gradient descent and an evolutionary algorithm known as a local search. They fundamentally solve the same problem in similar ways, but one has the advantage. Step-by-step they find a way down Loss Mountain. Watch real neural networks maximize the fitness of curve fitting. We've got Dogson here!~Timestamps~(0:00) Learning Learning(1:20) Neural Network Space(3:40) The Loss Landscape(7:21) The Blind Mountain Climber(8:37) Evolution (Local Search)(13:07) Gradient Descent(18:40) The Gradient Advantage(20:48) The Evolutionary (dis)advantageNuance for Nerds:Evolutionary algorithms have their place, and can be more performant than SGD in some circumstances, though it is uncommon. I stand by the accuracy of everything in the video, but probably should've added more caveats. SGD is not ALWAYS better than evolution, and I did not make this claim. SGD is generally accepted as the state of the art for optimizing the parameters of neural networks. Again, I will make a follow-up exploring more robust genetic algos and NEAT. I am a certified evolutionary algorithm turbo-fan.
Have you tried ChatGPT?s new free o3-Mini model yet? If not, you?re missing out on one of the most exciting AI tools available today. In this video, we?ll show you why the o3-Mini is mind-blowing and share 5 must-try prompts to get the most out of it.The o3-Mini is fast, efficient, and packed with features that make it perfect for everything from creative writing to problem-solving. Whether you?re a student, professional, or just curious about AI, this model is designed to help you work smarter and faster. We?ll walk you through how to use it and demonstrate its capabilities with real-world examples.By the end of this video, you?ll have a clear understanding of why the o3-Mini is a game-changer and how to use it to its full potential. Don?t miss out?discover the prompts that will unlock its power and transform how you interact with AI.What is ChatGPT?s o3-Mini model? How does it work? What are the best prompts to try? How can it help you? This video answers all these questions and more. Watch now to see why the o3-Mini is mind-blowing and how to use it like a pro!