The future of this is the same as what happened with PCs. The specificity will initially be thrown away, added back later as an accelerator, and eventually, brought back into the fold as an automatically used accelerator. It all comes full circle as optimizations, hinting, tiering, and heuristics.
AI will be used to select the net to automatically load. Nets will be cached, branch predicted etc.
The future of AI software and hardware doesn't yet support the scale we need for this type of generalized AI processor (think CPU but call it an AIPU.)
And no, GPUs aren't an AIPU, we can't even fit whole some of the largest models on these things without running them in pieces. They don't have a higher level language yet, like C, which would compile down to more specific actions after optimizations are borne (not PTX/LLVM/Cuda/OpenCL.)
It is believed that we evolved to taste bitterness because a lot of natural poisons are bitter. When people say something is an acquired taste, they are almost always referring to something bitter.
In many cultures bitter things are commonly eaten so people get past it even in childhood but in ones where they aren’t (like America) you find many adults who don’t enjoy bitter tastes.
I had been drinking bitter tea for quite a time. I don’t know if it has any health benefit but it kind of makes you addictive ;) Indeed, it is super bitter, half a leaf is enough for a full glass of hot water. However, after a while you will want to add more and more :D
I wonder if bitter food and drink can activate the bitter taste receptor as well. Maybe people who loves to eat bitter have a lower rate of cancer than the rest of us?
I'm a cilantro lover and it's just interesting how different my characterization probably is from yours. I wouldn't classify it as bitter in the least bit. To me, it's the purest taste of "freshness". I don't know how to better explain it.
There is an interesting genetic difference between cilantro-tolerant and cilantro-averse individuals. Similarly, some people smell asparagus in their pee, others don't. (The pee smells the same, in both cases.)
I read a long time ago, I think in Scientific American, that there are two different genes involved here: one that causes your pee to have that distinct asparagus smell or not, and a different gene that determines whether you can smell that particular aroma or not.
So the article divided people into four groups: smelly smellers, smelly non-smellers, non-smelly smellers, and non-smelly non-smellers.
I don't seem to be finding the original article, but here is a related one:
Part of my healthcare IT journey was as the LIMS Admin for 2 Tox labs, over the course of about 5 years. One of them was working on genetic markers for drug use. For example, of the 7 classes of Proton Pump Inhibitors, at that time, only 1 works for me. The rest either don't work, or work, but with lower efficacy.
The doc, handing it to me said, and I quote, "man, you got some weird genes". lol
We're getting close in our development of a spray that only attacks Cilandro plants. The current issue is the fire. Fire is the only way to truly kill Cilandro (other than nuking it from Orbit), but the fires tend to get out of hand, very, very fast.
As soon as we get over this final hurdle, you'll see the Cilandro of the world, ablaze and good riddance.
FFS, why is there even a type of soap that looks like a plant?
Best to write a launcher that simply forks an existing process that is in the right initial state. Other folks mention caching the ELF file but that will be slower than the loaded executable being cloned for a new process.
Unless you manually lock the executable in memory it will be evicted from RAM just like any other file on the filesystem. When you start up the usually-idle launcher it will load everything from disk (well load what it needs then the new process will load the rest as it starts up). I doubt the performance difference is significant unless your process does lots of work at startup and you will lose out on ASLR by always forking the same memory layout.
On Windows I believe one can call an API like VirtualProtect to prevent swap/eviction from RAM. Should be possible on Linux too?
Yeah, it'd only really be useful if one was running some type of web service that called an executable, and there is no source code available to do any fastCGI or whatnot.
Factory farmed meat is cancer too. I'd say, save the animals, and give a little toxicity to all the proles who haven't gone unprocessed plant-based yet.
Heh, you do realize that we live in a quantum supercomputer that computes at 10^50 Hz/Kg and 10^34 Hz/Joule?
The wave particle duality is just a min-decision/min-consciousness optimization.
Church-Turing thesis has no sign of being wrong - the maximum expressiveness of this universe is captured by computation.
The most complex theorems of the generalization of mathematics, computation, are actually about what would happen in formal systems, which physical systems are... So high complexity truth is... Simulcrums like Truman Show. Have fun, ahh
Consciousness is generated when the universe computes by executing conditionals/if statements. All machines are quantum/conscious in their degrees of freedom, even mechanical ones: https://youtu.be/mcedCEhdLk0?si=_ueWQvnW6HQUNxcm
The universe is a min-consciousness/min-decision optimized supercomputer. This is demonstrated by quantum eraser and double slit experiments. If a machine does not distinguish upon certain past histories of incoming information, those histories will be fed as a superposition, effectively avoiding having to compute the dependency. These optimizations run backwards, in a reverse dependency injection style algorithm, which gives credence to Wheeler-Feynman time-reversed absorber theory: https://en.wikipedia.org/wiki/Wheeler%E2%80%93Feynman_absorb...
Lower consciousnesses make decisions which are fed as signal to higher consciousnesses. In this way, units like the neocortex can make decisions that are part of a broad conscious zoo of less complex systems, while only being burdened by their specific conditionals to compute.
Because quantum is about information systems, not about particles. It's about machines. And consciousness has always been "hard" for the subject, because they are a computer (E) affixed to memory (Mc^2.) All mass-energy in this universe is neuromorphic, possessing both compute (spirit) and memory (stuff.) Energy is NOT fungible, as all energy is tagged with its entire history of interactions, in the low frequency perturbations clinging to its wave function, effectively weak and old entanglements.
Planck's constant is the cost of compute per unit energy, 10^34 Hz/Joule. By multiplying by c^2, (10^8)^2, we can get Bremmerman's limit, the cost of compute per unit mass, 10^50 Hz/Kg.
https://en.wikipedia.org/wiki/Bremermann%27s_limit
Humans are self-replicating biochemical decision engines. But no more conscious than other decision making entities. Now, sentience, and self-attention is a different story. But we should at the very least start with understanding that qualia are a mindscape of decision making. There is no such thing as conscious non-action. Consciousness is literally action in physics, energy revolving over time: https://en.wikipedia.org/wiki/Action_(physics)
Planck's constant is a measure of quantum action, which effectively is the cost of compute..or rather..the cost of consciousness.
Lines up a bit too perfectly. Everyone has their threshold of coincidence I suppose. I am working on some hard science into measuring the amount of computation actually happening, in a more specific quantity than hz, related to reversible boolean functions, possibly their continuous analogs.
AI will be used to select the net to automatically load. Nets will be cached, branch predicted etc.
The future of AI software and hardware doesn't yet support the scale we need for this type of generalized AI processor (think CPU but call it an AIPU.)
And no, GPUs aren't an AIPU, we can't even fit whole some of the largest models on these things without running them in pieces. They don't have a higher level language yet, like C, which would compile down to more specific actions after optimizations are borne (not PTX/LLVM/Cuda/OpenCL.)