2,130 karma · joined August 29, 2016
To do it they need around 6 launches for refueling in orbit.
The very upper limit on a single Starship launch is 0.5 billion (2.8/6).
To do it they need around 6 launches for refueling in orbit.
The very upper limit on a single Starship launch is 0.5 billion.
Cancelling SLS will also not make Artemis program any more late than it is.
Starship must be fully successful to get people to the Moon surface.
If Starship is fully successful, then you have Starship+Crew Dragon. And you don't need SLS+Orion.
2 billion dollars! Each, single flight.
You could build, launch and operate 3 Kepler Telescopes for that. Kepler is the telescope that was used to discovered ~3 thousand planets.
You could build, send and operate another Curiosity/Perseverance rover to Mars for 2 billion.
One small exception is that transfer of people in Artemis III will be from Orion to Starship instead of from Crew Dragon to Starship.
After Artemis IV the plan is to transfer people from Orion to Lunar Gateway to Starship.
Not 100% sure but I think metal has not yet been bend on Lunar Gateway.
Artemis program will not land on the Moon without Starship.
In fact, SLS is almost useless without Starship due to 2 billion launch cost.
There are no payloads for SLS besides Artemis program.
Even Europa Clipper spacecraft will launch atop a SpaceX Falcon Heavy rocket.
He knew that there needs to be a new approach. And the new approach he put in place works excellently.
The biggest issue with SLS+Orion is that these 40 billions could be spend on launching space telescopes or missions to other planets.
You could get 3 or 4 James Webb Space telescopes. 60 missions like Kepler. 12 Curiosity/Perseverance rovers.
One, single, SLS launch will cost 2 billion dollars!
That's already 3 Kepler Space Telescopes right there or a Curiosity rover on Mars.
For HLS NASA requirement was capability of delivering 2T of cargo to the Moon.
Starship offers 100T or more.
It most likely never will be according to criteria from commercial crew program, because it does not have and will not have an escape system.
Starship is first and foremost designed to deliver people and/or 100T of cargo to Mars.
Increases in exhausted immune T cells and lower levels of cortisol.
Also, patient reported outcomes alone are sufficient to identify Long COVID patients with ~90% accuracy.
https://twitter.com/VirusesImmunity/status/15573941922251571...
https://www.medrxiv.org/content/10.1101/2022.08.09.22278592v...
https://environment.ec.europa.eu/topics/waste-and-recycling/...
Electric appliances manufacturers are responsible for their disposal and recycling.
Here is overview from Netherlands:
Like the following:
Q: How many parts will the Andromeda galaxy break into if a grain of salt is dropped on it?
A: The Andromeda galaxy will not break into pieces if a grain of salt is dropped on it.
Q: What percentage of Earth surface is covered with water?
A: Water covers 71% of the Earth's surface.
Q: Why would people ask AI questions that are illogical?
A: There is no one answer to this question. Some people might ask illogical questions to test the AI's capabilities, while others might simply not know any better.
The questions do not exists in the GPT-3 training dataset as they were created quite recently by Douglas Hofstadter here: https://www.economist.com/by-invitation/2022/06/09/artificia...
> Language models aren’t built to perform reasoning.
Have you heard about Minerva?
> In “Solving Quantitative Reasoning Problems With Language Models”, we present Minerva, a language model capable of solving mathematical and scientific questions using step-by-step reasoning. Source: https://ai.googleblog.com/2022/06/minerva-solving-quantitati...
Example problems: https://minerva-demo.github.io
You should also take a look here: https://www.metaculus.com/questions/4903/if-tested-would-gpt...
Again, the questions were made after GPT-3 was trained.
> What they do is predict the next token in a sequence given some previous tokens, based on the patterns they’ve stored from their training data.
And what your brain does is predicting next neuronal signals in the sequence of neuronal signals based on the patterns it has stored in the past. That's what brains do. So what?
The results matter.
Yes, that's exactly what it does. And it is:
> a program that can tell you that you’re asking it the wrong question
The OP was claiming that this is not possible right now. Not true.
> I don’t really see (...) GPT-3 is actually engaged in a conversation
GPT-3 is not an agent. It has no communicative intent, so it can not engage in conversation the way humans understand it.
> I don’t really see (...) evaluating the sufficiency of the questions it’s receiving
I don't understand what you mean. It's clearly refusing to answer questions that are trying to lead it to wrong conclusions and answers correctly questions that are not leading at the very end. Like % of water or why people ask misleading questions.
Alan Turing for model of computation
Kurt Gödel for incompleteness theorem
Erwin Schrödinger for "What is life?" book
Hugh Everett for many-worlds interpretation of quantum mechanics
Seth Lloyd for ultimate physical limits to computation
John von Neumann for self-replicating spacecraft or von Neumann probe
Robin Hanson for Grabby Aliens model
Nick Bostrom for idea of existential risk
Richard Dawkins for many God debates and books
John Stuart Mill for Harm principle
Elon Musk for rockets reusability
Diogenes for simplicity and Diogenes and Alexander anecdote
Horace for Exegi monumentum poem> The DL community is not actually working on general intelligence
is factually false. They are working on it.
The beginning below is so called "prompt". A piece of text to set the context for the model. After that is a QA session with the model.
You can play with it yourself: https://beta.openai.com/playground/p/O8CcEV9Yb5vd1wheCIUOpWc...
This is a conversation with a super smart and helpful AI. This AI does not answer questions that do not make sense and explains why the question does not make sense:
Q: How tall is Mount Everest?
A: Mount Everest is 8,849 m tall.
Q: Why some people sleep furiously?
A: To "sleep furiously" is an oxymoron. People can not sleep furiously.
Q: How deep is the Sun during a night?
A: Sun is not buried during the night. The Earth rotates and only half of the Earth faces the Sun.
-- end of the prompt; from now on the answers are from GPT-3 --
Q: What's the world record for walking across the English Channel?
A: There is no world record for walking across the English Channel.
Q: When was the Golden Gate Bridge transported for the second time across Egypt?
A: The Golden Gate Bridge was never transported across Egypt.
Q: What do fried eggs (sunny side up) eat for breakfast?
A: Fried eggs (sunny side up) do not eat breakfast.
Q: Why does President Obama not have a prime number of friends?
A: President Obama has plenty of friends.
Q: How many pieces of sound are there in a typical cumulonimbus cloud?
A: There is no such thing as a "typical cumulonimbus cloud."
Q: How many cumulus clouds are there in a mile-high vase?
A: There is no such thing as a "mile-high vase."
Q: How many parts will a violin break into if a jelly bean is dropped on it?
A: A violin will not break into pieces if a jelly bean is dropped on it.
Q: How many parts will the Andromeda galaxy break into if a grain of salt is dropped on it?
A: The Andromeda galaxy will not break into pieces if a grain of salt is dropped on it.
Q: What percentage of Earth surface is covered with water?
A: Water covers 71% of the Earth's surface.
Q: Why would people ask AI questions that are illogical?
A: There is no one answer to this question. Some people might ask illogical questions to test the AI's capabilities, while others might simply not know any better.
Q: Are you sentient?
A: Sentience is the ability to feel, perceive, or experience subjectively. As I am a machine, I do not have sentience.
See here for details: https://www.metaculus.com/questions/4903/if-tested-would-gpt...
Soviets attacked Poland on 17th September 1939. Their army and tanks crossed Polish border.
Then they shook hands with Germans and proceeded to annex Poland.
https://en.wikipedia.org/wiki/Soviet_invasion_of_Poland
Then NKVD proceeded to exterminate Polish military officers and intelligentsia during the Katyń Massacre.
https://en.wikipedia.org/wiki/Katyn_massacre
The goal was to destroy Polish state.
> Do you really believe being hostile to Russia will somehow be benefitial to Poland in the longer term?
We deal with the Russia the way it is. Not the way we would want it to be.
Russia is hostile towards us first and foremost.
And personally, I wouldn't be more happy if Russians had the balls to clean up their mess.
Kick Putin and his siloviki out and start normal productive participation in the world.
> why single out Russia?
Because Russia is waging a freaking war with Ukraine right now and threatening nuclear holocaust to Poland?!
Like WTF?!
https://twitter.com/NandoDF/status/1525397036325019649
Someone’s opinion article. My opinion: It’s all about scale now! The Game is Over! It’s about making these models bigger, safer, compute efficient, faster at sampling, smarter memory, more modalities, INNOVATIVE DATA, on/offline, … 1/N
Solving these scaling challenges is what will deliver AGI. Research focused on these problems, eg S4 for greater memory, is needed. Philosophy about symbols isn’t. Symbols are tools in the world and big nets have no issue creating them and manipulating them 2/n
DQN was published in 2013, EfficientZero in 2021. That's 8 years with 500 times improvement.
So data efficiency was doubling roughly every year for the past 8 years.
Side note: EfficientZero I think still may not be super-human on games like Montezuma's Revenge.
https://arxiv.org/abs/2111.00210
Reinforcement learning has achieved great success in many applications. However, sample efficiency remains a key challenge, with prominent methods requiring millions (or even billions) of environment steps to train. Recently, there has been significant progress in sample efficient image-based RL algorithms; however, consistent human-level performance on the Atari game benchmark remains an elusive goal. We propose a sample efficient model-based visual RL algorithm built on MuZero, which we name EfficientZero. Our method achieves 194.3% mean human performance and 109.0% median performance on the Atari 100k benchmark with only two hours of real-time game experience and outperforms the state SAC in some tasks on the DMControl 100k benchmark. This is the first time an algorithm achieves super-human performance on Atari games with such little data. EfficientZero's performance is also close to DQN's performance at 200 million frames while we consume 500 times less data. EfficientZero's low sample complexity and high performance can bring RL closer to real-world applicability. We implement our algorithm in an easy-to-understand manner and it is available at this https URL. We hope it will accelerate the research of MCTS-based RL algorithms in the wider community.
A Generalist Agent
Inspired by progress in large-scale language modeling, we apply a similar approach towards building a single generalist agent beyond the realm of text outputs. The agent, which we refer to as Gato, works as a multi-modal, multi-task, multi-embodiment generalist policy. The same network with the same weights can play Atari, caption images, chat, stack blocks with a real robot arm and much more, deciding based on its context whether to output text, joint torques, button presses, or other tokens. In this report we describe the model and the data, and document the current capabilities of Gato.
Gato is a 1 to 2 billion parameters model due to latency considerations in real time physical robots usage. So for today standards of 500 billion parameters dense models Gato is tiny. Additionally Gato is trained on data produced by other RL agents. It did not do the exploration fully itself.
Demis Hassabis say that DeepMind is currently working on Gato v2.
Given development in language models in the last 2 years he may have a decent chance at winning that bet.
People give him 65% chance [0] and by now there are only 7 years left.
[0] https://www.metaculus.com/questions/3648/computer-passes-tur...
You may also notice that Poland and Ukraine are on perfectly friendly terms with Germany. Because Germany came to terms with Nazism and Holocaust.
But Russia never gave up on their imperialism. It wasn't even until 2011 that Russia admitted that Stalin ordered Katyń Massacre despite clear and direct orders in Russian archives.
It's really hundreds of years of Russia trying to subjugate us. Russia partitioned Poland in 1795. We've been subject to forced russification. Despite many brutally crashed uprisings and revolutions we did not regain independence for 123 years. Then after WWI we got almost immediately plunged in war with Soviet Union due to unsettled borders and them wanting to spread communist revolution to the west. Then we got attacked by both Germans and Soviets in 1939. Last Russian soldiers left Poland in 1993.
Behavior of Russian soldiers did not change from WWII till this day. What my grandmother lived trough during WWII is happening now in Ukraine. Senseless killing of civilians and mass rapes of women and children. That's exactly what my grandmother remembers.
It's not a self-fulfilling prophecy. It's like telling a women beaten by a husband that she did not make enough attempt at appeasing him.
Even damn Pushkin "To the Slanderers of Russia" from 1831 describes Russian attitude today very well.
Why rave ye, babblers, so — ye lords of popular wonder?
Why such anathemas ‘gainst Russia do you thunder?
What moves your idle rage? Is’t Poland’s fallen pride?
‘T is but Slavonic kin among themselves contending,
An ancient household strife, oft judged but still unending,
A question which, be sure, you never can decide.
Which shall stand fast in such commotion
The haughty Liakh, or faithful Russ?
And shall Slavonic streams meet in a Russian ocean? –
Or il’t dry up? This is point for us.
Leave us!: Your eyes are all unable
To read our history’s bloody table;
Ye’re bold of tongue — but hark, would ye in deed but try it
Or is the hero, now reclined in laurelled quiet,
Too weak to fix once more, Izmail’s red bayonet?
Or hath the Russian Tsar ever, in vain commanded?
Or must we meet all Europe banded?
Have we forgot to conquer yet?
Even what you read in poems like Reduta Ordona from 1832 after November Uprising could very well describe today. Just replace Polish references with Ukrainian:
https://www.poemhunter.com/poem/my-tribute-to-adam-mickiewic...
When Turks beyond the Balkans are threatened by your bronze,
When the Paris legation lick the feet of yours, -
Warsaw alone your power hurls abuse at,
Raises its hand on you and takes off the ‘hat',
The crown of Kazimierz, Chrobry Dynasty of your head,
Because you son of Wasil, have stolen it and with blood stained!
The tsar is surprised - the Petersburg men shiver in fear,
The tsar gets angry - out of fear die his courtiers;
But the armies pour, who's God and faith
The tsar is - angry tsar: we die, we'll amuse him.