Quantum Time (Andy Mendelsohn)

81 posts

Quantum Time (Andy Mendelsohn)

Quantum Time (Andy Mendelsohn)

@quantumtime

Time in quantum theory appears irreversible, yet delayed quantum eraser expt may work by backwards propagation in time or not.

Tham gia Haziran 2009
1.3K Đang theo dõi68 Người theo dõi
Terry Kim
Terry Kim@thewayofkaizen·
This was the most brutal insult Steve Jobs ever received: In 1997, a man publicly insulted Steve Jobs at the Worldwide Developer Conference in front of thousands. Jobs' response changed Silicon Valley forever.. Here's what happened (and how this defined his legacy forever):🧵
Terry Kim tweet media
English
100
772
4.9K
3M
Quantum Time (Andy Mendelsohn) đã retweet
Whitehead Institute
Whitehead Institute@WhiteheadInst·
Many chronic diseases have a common denominator that could be driving their dysfunction: reduced protein mobility, which in turn reduces protein function. A new paper from @youngricka's lab describes this pervasive mobility defect: wi.mit.edu/news/cellular-…
Whitehead Institute tweet media
English
3
15
20
4.2K
Quantum Time (Andy Mendelsohn)
Quantum Time (Andy Mendelsohn)@quantumtime·
@CellRepJrnl @eskamoah @cc_biology_ @rejuvenatebio Both promising that there was some life extension and disappointing in that the modest effect shows either that partial of epigenetic age was too partial, perhaps due to technical reasons, or that epigenetic age reversal is not the whole story when it comes to rejuvenation.
English
0
0
0
184
Rowan Cheung
Rowan Cheung@rowancheung·
Gemini 1.5 Pro's 1,000,000+ token context length is incredible. I got early access and spent my Saturday night running tests. Here are 6 impressive capabilities I found:
Rowan Cheung tweet media
English
135
658
5.3K
2.6M
Quantum Time (Andy Mendelsohn)
Bitcoin is engineered to have an artificially limited supply on the one hand, but is easily clonable given its open source nature on the other. That such clones haven't gained traction is just a matter of hype and fashion. The extreme volatility of bitcoin makes it an unsuitable world currency, which could destabilize global economic activity.
English
0
0
0
39
Brian Armstrong
Brian Armstrong@brian_armstrong·
One idea I’ve been contemplating is that Bitcoin may be the key to extending western civilization. The natural trend of whichever country has the reserve currency is to inflate the money supply and increase deficit spending until it loses that advantage. The U.S. is somewhere on this journey, which many have discussed including Dalio in The Changing World Order. The Yuan and Euro have their own issues and aren’t viable alternatives currently, so the assumption is the U.S. can continue to inflate, but what I think many haven’t considered is that people have an alternative now with crypto. They may start moving fiat into crypto, as an antidote to inflation. Contrary to what some may assume, I don’t think this be a threat to the dollar and the U.S., I think it will be a natural check and balance that will complement the dollar, and be the best defender of long term American interests (and western civilization more broadly). It’s better to move from dollars to crypto than to another country or region’s fiat. I also think both fiat and crypto will co-exist for a long time. They are more complements than substitutes. And dollar backed stable coins like USDC, or flat coins, will play a major role in unifying these worlds. Still thinking this through. What other good takes are there on this?
English
1.1K
1.3K
8.9K
2.2M
Quantum Time (Andy Mendelsohn)
Quantum Time (Andy Mendelsohn)@quantumtime·
@AgingBiology Predictions about such extended human longevity should probably be considered highly speculative. OTOH there is great potential to reverse aging at the epigenetic level.
English
0
0
0
70
Quantum Time (Andy Mendelsohn)
Quantum Time (Andy Mendelsohn)@quantumtime·
@migtissera You forgot 2020: AlphaFold which allows prediction of protein structure, winning the 2023 Lasker Prize for DeepMind inventors Hassabis and Jumper, but really for AI.
English
0
0
1
368
Migel Tissera
Migel Tissera@migtissera·
This may be a hot take, but I'm willing to look at both sides. 2012: AlexNet -- Kickstarted the whole fucking Deep Learning movement. First work ever in the world to use CUDA to accelerate neural networks. NVIDIA is in AI from here. 2014: Dropout -- A highly efficient way to regularize a deep learning network. 2014: Sequence-to-Sequence Learning -- Kickstarted language translation. 2016: TensorFlow -- One of the best Deep Learning frameworks 2016: AlphaGo -- First ever software system to surpass human performance in the game Go. 2019: OpenAI-Five: DOTA-2 -- Reinforcement Learning algorithm beating world's best DOTA-2 players. 2021: Codex -- Large Language Model for Code 2017 - Present: GPT-1, GPT-2, GPT-3, GPT-3.5, GPT-4 -- World class language models that, in many cases, surpass human abilities. 2023: GPT-4-V -- World's best multi-modal language and vision model Just look at the contributions by this guy (@ilyasut) before you put him in either camp (EA or e/acc). Me personally, I'd like to see acceleration, and I frequently release my work Open Source. Boardroom politics is shit, I don't think Ilya liked doing it either. But if he thinks OpenAI was moving too fast, we probably should listen to him given his contributions.
English
33
98
847
318K
Quantum Time (Andy Mendelsohn)
@Cal_Engineer This is an underappreciated paper so far-- certainly quite consequential regarding earlier DNA clock-based publications. Would be useful to reevaluate raw data from those using Conboy's new model.
English
0
0
0
9
Berkeley Engineering
Berkeley Engineering@Cal_Engineer·
Research from the Conboy Lab was featured on the cover of the journal “Aging.” Led by Irina Conboy, professor of bioengineering, the lab’s exciting new research centers on the measurement of biological age. View the article: bit.ly/3Pqe8Bn
Berkeley Engineering tweet media
English
1
0
2
431
Quantum Time (Andy Mendelsohn)
Quantum Time (Andy Mendelsohn)@quantumtime·
@prof_horvath How can these results be reconciled with David Sinclair's recent study that presented evidence that ds DNA breaks could drive epigenetic aging (such as DNA methylation clocks)?
English
0
0
0
409
Quantum Time (Andy Mendelsohn)
Quantum Time (Andy Mendelsohn)@quantumtime·
Focusing on the ageing brain, they show that "activation of STING triggers reactive microglial transcriptional states, neurodegeneration and cognitive decline."
English
0
0
0
47
Quantum Time (Andy Mendelsohn)
Quantum Time (Andy Mendelsohn)@quantumtime·
Nature: cGAS/STING play a major role in inflammaging. The author's show that "cGAS–STING signalling pathway, which mediates immune sensing of DNA, is a critical driver of chronic inflammation and functional decline during ageing." nature.com/articles/s4158…
English
1
0
0
107
Peter Adams🇪🇺🇺🇳
Peter Adams🇪🇺🇺🇳@AdamsBioAging·
Excited to be part of this. Scientists take note - you can set up your own journal. The big publishers are not indispensable. Aging Biology is by scientists, for scientists.
Agingdoc🩺Dr David Barzilai🔔MD PhD MS MBA DipABLM@agingdoc1

Delighted to share the news @AgingBiology01 👀 Aging Biology is now publishing papers & accepting submissions!👇 🔗agingcelljournal.org Congratulations @AdamsBioAging and fellow EICs Vera Gorbunova, John Sedivy, Julie Andersen! 🎉

English
4
25
124
20.6K
Quantum Time (Andy Mendelsohn)
Quantum Time (Andy Mendelsohn)@quantumtime·
@stackhodler The Bitcoin bubble is the consequence of creating a system appealing to insatiable greed: synthesize money out of the ether, and then make it increasingly more difficult to do so with time. Just needs publicity and herd mentality to work. Appeals to humanity's basest instincts.
English
1
0
1
78
Stack Hodler
Stack Hodler@stackhodler·
"The bursting of the Bitcoin bubble" - The Economist, October 2011 People have been calling Bitcoin a bubble since it was worth $30 in 2011. 12 years later it's worth $30,000. If you're still calling it a bubble or a fad, maybe now is the time to ask why you've been wrong so far. Ignore the price for a moment. What is Bitcoin actually? It's the solution to a global problem: How to store wealth long-term without the risk of debasement and seizure. It's is a technological breakthrough. Teslas are engineered to be the best car. iPhones are engineered to be the best mobile device. Bitcoin is engineered to be the best store of wealth. Do you think investing in tech stocks is normal but investing in a technology that addresses one of the biggest problem on earth (storing value) is dumb? 🤔 The majority of people are storing their wealth in assets that leak value to inflation or debasement over time and are liable to seizure by increasingly capricious governments. Sovereign debt, equities, real estate, gold, fine art... They are all imperfect stores of wealth, but they were the best we had until Bitcoin. "But Bitcoin has no intrinsic value!" Here's why Bitcoin is valuable: It's Finite: There are only 21 million coins. There's no central banker who can debase your wealth. It's Divisible: you can buy / sell $0.10 worth, or $10 Billion worth. It Can't Be Seized: It can't be physically taken from you. There are ways to store it so that no external party can take it from you. You have full control over your property. Do you know anything else like that? It's Supply Inelastic: Unlike almost everything else on earth, as the price of Bitcoin rises, there's no way to make more of it. That prevents the price from falling due to a sudden supply glut. It's Portable: You can move anywhere in the world and have access to your Bitcoin without permission. It weighs nothing and you can move across borders without anyone knowing you own it. It's Free From Counter-party Risk: The entire financial system could melt down around you. JP Morgan could go bankrupt. And you could still have your coins without needing to be bailed out by anyone. It's Globally Accessible: Storing wealth is a global problem. And Bitcoin isn't jurisdiction specific. No one gets special treatment - it's a fair protocol available to all. No other asset on earth has these combined characteristics. It had to be invented. #Bitcoin was engineered as the ideal store of value. It has gained more since March 2020 than the S&P500 has gained since the bottom of 2009. It's not just some widget that anonymous people like me are hallucinating over. Now that Blackrock, Fidelity, Deutsche Bank, Crédit Agricole, Citadel, and other large institutions are about to offer it to their clients you can either dig your heels in and say that they're just getting in on the scam, or you can try to understand why 70% of supply hasn't been sold in over a year despite large volatility. Why are Billionaires stacking Bitcoin? Why is famous trader Paul Tudor Jones saying he's "never sat on a horse as long as he's sat on Bitcoin?" Are we all morons? I certainly don't feel like a moron watching investments 10x over a 4 year period. The time to ask questions with an open mind is running out. The institutions are coming. And the fiat wave won't be far behind them. The window of opportunity on the most obvious investment of our generation is closing. You'll still benefit by stacking BTC at $300K a coin. But you'll hate yourself for ignoring this tweet.
Stack Hodler tweet media
English
247
985
4.1K
906.2K
Quantum Time (Andy Mendelsohn)
Quantum Time (Andy Mendelsohn)@quantumtime·
@itsPaulAi Beware!!! Just tried a version of this on a search related to frontotemporal dementia and Bard like ChatGPT fabricated references. I even added seatch Google Scholar and it still persisted in this.
English
0
0
0
87
Paul Couvert
Paul Couvert@itsPaulAi·
Prompt → I am writing a thesis on "Economic relations between the West and the Middle East". Give me a table with academic references that are less than 3 years old with the title, author, date and a short summary.
English
10
27
255
73.8K
Paul Couvert
Paul Couvert@itsPaulAi·
Don't use ChatGPT for academic research. Instead use Bard, the free alternative from Google. Here is how to use Bard smartly for academic purposes:
English
271
1.8K
8.4K
1.8M
Quantum Time (Andy Mendelsohn)
@elonmusk Give us the ability to increase the frequency of seeing certain topics and people by creating a rating scale for users. No need to make it public, just have twitter use it internally.
English
0
0
0
7
Elon Musk
Elon Musk@elonmusk·
What should do more of or less of or aren’t including at all?
English
15.9K
1.3K
16.9K
8M
Elon Musk
Elon Musk@elonmusk·
We’re trying hard to make your feed as compelling as possible (maximize unregretted user minutes). How is it now vs 6 months ago?
English
14.8K
10.8K
46.8K
31.7M
Quantum Time (Andy Mendelsohn)
@ylecun True, OTOH there will be benefits to allowing AI free rein in terms of greater creativity if nothing else. Therefore there is and will be motivation to create such entities.
English
0
0
0
1
Robert F. Kennedy Jr
Robert F. Kennedy Jr@RobertKennedyJr·
Ellen Brown makes a strong case the FDIC/SEC war on crypto caused failures of SVB, Signature, and Silvergate banks. FDIC and SEC have no authority to wage an extra-legal war on crypto that leaves major banks as collateral damage. #Kennedy24 ellenbrown.com/2023/04/29/how…
English
417
1.8K
9.4K
1.1M
Quantum Time (Andy Mendelsohn)
@UrnovFyodor Good idea, but it would need to have a near complete set of papers as a database. Currently, perhaps only a few organizations (Google?, a few of the biggest universities) have sufficient access to create the database.
English
1
0
0
759
Fyodor Urnov
Fyodor Urnov@UrnovFyodor·
I wonder if ChatGPT could be used to ask this question during peer review: "Has one key finding of this work been published before?" In this case the answer would be "yes" (in 2007). Clearly we need AI since "human I" appears to have underperformed here. #smh
Fyodor Urnov tweet media
English
8
17
105
53.5K
Quantum Time (Andy Mendelsohn)
@heykahn Actually the technology underlying ChatGPT will replace many jobs in addition to saving people work. Think about it this way, if you are correct, and 100's of hours of work are saved for each person, then not as many people are needed at a workplace.
English
0
0
0
149
Zain Kahn
Zain Kahn@heykahn·
ChatGPT isn't going to replace your job. It's going to replace the most boring parts of your job. Top 10 prompts to save 100s of hours of boring work:
English
117
1K
6.6K
1.7M
Elon Musk
Elon Musk@elonmusk·
@Scobleizer Maybe we are LLMs, living on a GPU & don’t realize it 👀
English
1.7K
782
10.9K
2.3M
Robert Scoble
Robert Scoble@Scobleizer·
I just realized if @elonmusk wants TruthGPT then he will have to build the best AI validator. Then I realized Twitter is the best validator of all. Think about it. GPT spits out a list of facts. How would you best validate? A Twitter Space. Talking with other humans. That could be quickly ingested and add on meta data to the topic at hand. Let’s say we were discussing plug-ins. Couldn’t the humans figure out whether there is some bullshit being generated? Yup. Feed that audio into an LLM and the world’s best validator will be born. Add Tesla data and it will answer questions no one else can. “I see a fire on The Golden Gate Bridge. Can you tell me what is burning?” Freshly validated every 10 seconds as another Tesla drives by with its computer vision system running. Then someone Tweets a video. Validated again. Every business and every human will want a validator. Elon has the ability to build the best.
Chattanooga, TN 🇺🇸 English
336
305
3.2K
1.7M