Show Posts
This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
Zuckerberg's STUNNING Statement: ?AI Will Write MOST Software Soon"
Google TITANS and Sakana Transformer?. AI gets memory and can keep learning after training. #ai #ainews #agi
Titans: Learning to Memorize at Test Time https://arxiv.org/abs/2501.00663
Sakana Transformer? https://sakana.ai/transformer-squared/
0:00 Intro
0:34 Google Titans architecture
1:44 Transformers model and limitations
5:14 Giving AI memory
6:35 AI training vs test time
8:57 Titans long term memory
12:59 Titans architecture
19:50 Pros and cons
24:00 Performance of Titans models vs Transformers
27:31 Sakana Transformer2
30:50 Model design and rationale
32:24 How Transformer2 works
36:00 Transformer2 performance
38:12 Conclusion
Here's a modified version of the search problem above, from the search space side. Guess a number between 0 and 2^n. Again, the verification process is easy. But the solution would take log(2^n) = n steps.The size of the problem can be made to look deceptively small. For example, binary search for a natural number up to googol needs up to 333 steps, because a googol requires 333 bits to express in binary. When the upper limit is increased to a googolplex, the number of steps to solve it increases to around 3.32x10^100.
It would be even longer to guess a number between 0 and n^n.
Were you aware of this?Not the exact formula. But my experiments with microwave and metamaterials suggest that refractive index depends on the unit density of metamaterial and polarizability of the metamaterial unit.
https://en.wikipedia.org/wiki/Clausius%E2%80%93Mossotti_relation
Why? It's pretty obvious that a conductor carrying a couple of amps at 50 - 60 Hz is going to radiate em. The question is whether that actually matters in the context of all the other current carriers in the house.It depends on the shape of the conductors, whether it will act like an antenna, or more like a Faraday cage.
LED lamps obviously emit less because they are DC-powered, but if you obtain that power from the mains, you are probably emitting 100 kHz from the switch-mode power supply.That oscillation can be radiated to its surroundings. It can affect the temperature measurement of some types of thermometers, especially the electronic ones.
What makes you fail to distinguish between impedance and refractive index?Most of us know that impedance is frequency dependent. Mathematical model for impedance is a complex number. Thermodynamics and black body radiation seem to consider only the real part of this impedance. That's because in those cases, the imaginary part is not significant. They're only special cases from the more general cases where the radiating or radiated bodies are not necessarily black body.You have that more or less exactly wrong.
It's understandable that scientific research started with simpler cases. But let's not forget that there are more than what earlier discovery has told us.
The imaginary part of the refractive index relates to absorbing light.(And thus BBR) But the real part relates to the wavelike properties of the light that carries on or is reflected.
"Light propagation in absorbing materials can be described using a complex-valued refractive index.[2] The imaginary part then handles the attenuation, while the real part accounts for refraction"
From
https://en.wikipedia.org/wiki/Refractive_index
Did you consider learning science (rather than just watching YT)?
Stargate is a $500 billion AI infrastructure project launched by Donald Trump, OpenAI, SoftBank, and Oracle to build the largest AI data centers and supercomputers in the world. Designed to strengthen America's leadership in artificial intelligence, it aims to outpace China and revolutionize industries like healthcare, finance, and robotics. This ambitious initiative promises massive job creation, advanced AI technology, and a transformative impact on the global tech landscape.
Key Topics:
Stargate, the $500 billion AI infrastructure project backed by Trump, OpenAI, and SoftBank
The construction of massive data centers and supercomputers to power advanced AI technologies
How Stargate aims to outpace China in AI development and revolutionize industries like healthcare and robotics
What You?ll Learn:
Why Stargate is the most ambitious AI project in history and what it means for the future
How this initiative is set to transform industries and create cutting-edge AI capabilities
The role of companies like Oracle, Nvidia, and Microsoft in shaping the AI landscape
Why It Matters:
This video uncovers the groundbreaking Stargate project, exploring its potential to redefine global AI leadership, drive innovation, and secure America?s dominance in the rapidly evolving tech world.
Nvidia has done it again! The groundbreaking release of NVLM 1.0 is sending shockwaves through the tech world. But what exactly makes this new technology so revolutionary? In this video, we?ll explore why NVLM 1.0 is set to change the game for AI, machine learning, and the entire tech industry.
From its unparalleled processing power to its cutting-edge AI capabilities, NVLM 1.0 is designed to push the boundaries of what's possible. We?ll break down the key features that make it a must-have for developers, researchers, and tech enthusiasts alike.
Find out how Nvidia?s latest innovation is redefining efficiency, speed, and scalability in ways the industry has never seen before. Whether you?re into AI development, high-performance computing, or future-forward tech, this is one release you can?t afford to ignore.
What makes NVLM 1.0 stand out? How will it transform industries beyond AI? Can Nvidia maintain its lead in the ever-competitive tech space? Watch now to find out why everyone is talking about Nvidia?s newest game-changer!
Google DeepMind is building a groundbreaking AI system capable of simulating the entire physical world to advance toward Artificial General Intelligence (AGI). By combining multimodal data like video, audio, and robotics, this world simulation AI aims to replicate real-world physics for applications in robotics, gaming, and scientific research. This ambitious project highlights Google's focus on scaling AI models to achieve unprecedented levels of intelligence and realism.
Key Topics:
Google?s groundbreaking AI initiative to simulate the physical world for AGI development
The integration of multimodal data like video, audio, and robotics in world simulation
Real-world applications of AI-driven simulations in robotics, gaming, and scientific research
What You?ll Learn:
How Google?s AI is paving the way to Artificial General Intelligence through world modeling
The transformative potential of AI simulations in training robots and creating realistic environments
Why these advancements represent a critical leap in technology and intelligence
Why It Matters:
This video explores the revolutionary concept of world simulation AI, delving into its potential to reshape industries, advance technology, and bring us closer to achieving true AGI.
According to Planck and Einstein, E = hf. Where does your "n" come from?I asked Gemini your question.
An electron moving at constant velocity does not emit emr.
The formula E = n * h * f comes from the concept of quantization of energy, a cornerstone of quantum mechanics.
Here's the breakdown:
* Planck's Hypothesis: In 1900, Max Planck proposed that energy is not emitted or absorbed continuously, but rather in discrete packets called quanta. These quanta of energy are now known as photons.
* Planck's Constant: To describe this quantization, Planck introduced a fundamental constant, 'h', now known as Planck's constant. It has a value of approximately 6.626 x 10^-34 joule-seconds (J⋅s).
* Energy of a Single Photon: Planck hypothesized that the energy (E) of a single photon is directly proportional to its frequency (f). Mathematically, this relationship is expressed as:
* E = h * f
* Multiple Photons: If you have 'n' number of photons, the total energy (E) would be:
* E = n * h * f
In simpler terms:
Imagine light as a stream of tiny energy packets. Each packet's energy is determined by its frequency (how fast it oscillates). Planck's constant (h) is a conversion factor that relates the frequency of the light to the energy of each packet.
Key Points:
* This equation revolutionized our understanding of light and laid the foundation for quantum mechanics.
* It has numerous applications, from explaining the photoelectric effect to understanding the behavior of atoms and molecules.
Big deal. I have never ever wanted to work, i just had to and even sometimes i had to say that i liked my work. Now i really dont care anymore. I retired 3 years ago. Though at the same time i really do not envy young people. Now i do not know what my grandkids are gonna be and do. Kinda scary and at the same time fascinating. Humans have that feeling when we are facing something unknown.
Jobs aren't going anywhere, it will just be a matter of whether they are paid or not. Who knows what economic model we will end up with. I just hope it's not the current capitalist system we have now or we're all screwed, and I say this as a capitalist myself. AI will initially just make existing employees more efficient, but when we reach true AGI why would companies need to hire people at all? Tech always gets cheaper, faster, and more efficient. When people have to compete with AI for labor salaries are going to go south like a duck in winter. Even before that we will see a lowering of entry barriers for jobs and previously high paid positions will be filled by everyone looking to make more money. This will cause salaries to start to average across the board to a certain degree. Unfortunately, your debt will stay the same, so good luck paying off that $500k home when you make 1/10th as much. Do people think corporations will share the enormous wealth they are about to receive? They certainly haven't during this period of massive inflation. Corporations have reported record profits, but wages have stayed stagnant.
So in your opinion everything that evolves is conscious?Consciousness is not a binary quality.The effects of evolutionary process in an environment can be seen more clearly by contrasting with conditions without evolutionary process. The latter will be extremely less likely to cause improvements in survival.Even if the environment doesn't change, and if the things that live in it are not the first living things ever, they must have evolved. I cannot imagine any environment in which evolution has not happened. With the possible exception of the United States presidency.
DeepSeek Just released their first reasoning model called Deepseek-R1 that is on par with Openai O1 reasoning model. We will have a quick look at the release in this video.
DeepSeek R1 is probably the best open weight reasoning model to date. I tested it on coding and reasoning tasks. The performance is pretty impressive.
Based on statistics, human drivers are not that reliable either. And soon they will be generally worse than standard autonomous vehicles.So autonomous vehicles are good news for lawyers, but not for anyone inside or outside the vehicle.Quotefrom: alancalverd on 13/01/2025 10:22:04It depends on several factors.
Same old question: if an autonomous vehicle kills someone, who is liable?
Unfortunately the makers of infrared heat lamps do not share your profound knowledge!A good absorber is also a good emitter in the same spectrum. That's where infrared heat lamp can optimize it.
Like sleeping or going to a vacation.People need some break from routine activities.Such as kneeling before a parasite every 7 days?
Only conscious entities can have a goal or objective. Until then, the improvement in survival can be considered as merely side effects.In the perspective of the survivors, being better at survival is preferred. Their perspective is what matters.Time doesn't flow backwards. Neither does evolution. The test of survival cannot precede the evolution of the object being tested. So evolution cannot have an objective.