Discover and read the best of Twitter Threads about #computation

Most recents (14)

[1/12] 🧵 @Conste11ation ( $DAG) is a highly scalable and secure distributed ledger technology (DLT) that achieves zero/low trust interaction platforms, peer-to-peer (P2P) internet overlay protocols, and platform-neutral computation description language. Let's dive in! 🚀
[2/12] 🔐Zero/Low Trust Interaction Platforms (Shared Security):
@Conste11ation creates a shared security environment through its L0 consensus mechanism. L0 allows metagraphs (custom consensus apps) to maintain their own L1 layer while sharing L0 security with the global network.
[3/12] 🌐This shared security model reduces the need for trust as metagraphs inherit the security of the global L0 network. It's similar to Polkadot's shared security model, where parachains share security provided by the relay chain. #cryptocurrency #DAG
Read 16 tweets
"Liquid and Solid #Brains: Mapping the #Cognition Space"

Today's SFI Seminar by Ext Prof @ricard_sole, streaming now — follow this 🧵 for highlights:

Image
"Why #brains? Brains are very costly...it seems like they are not a very good idea to bring complex cognition to a #biosphere that just needs simple replicators."

"I also want to explore the problem of #consciousness, which is around all the time..."

@ricard_sole Image
Two different views of #evolution:

- Stephen Jay Gould and an emphasis on #contingency

- Simon Conway Morris/Pere Albach and an emphasis on #constraints and #convergence

--> "Replay the tape with different results vs. 'the logic of monsters' & 'life's solution'" Image
Read 21 tweets
Decision trees based Machine Learning models are some of the best performant algorithms in eras of predictive capability, especially on small and heterogenous datasets.

1/4
They also provide an unparalleled level of interpretability compared to all other non-linear algorithms. However, they are very hard to optimize on Von Neumann architecture machines due to their non-uniform memory access patterns.

2/4
In groundbreaking work published in Nature Communications a team of researchers has shown that analog content addressable memory (CAM) devices with in-memory calculation can dramatically accelerate tree-based model inference, as much as 10**3 over the conventional approaches 3/4
Read 4 tweets
Fees on #StarkNet
The idea is to compute the L1 footprint of a transaction, including 2 dimensions:
- computational complexity: the heavier the transaction, the larger its portion in the proof verification cost
- on chain data: L1 call data cost
Let see how it works 🧵
Users can specify the maximum fee they are willing to pay via max_fee field of the transaction.
Fee mechanism:
#Computation =
Number of Cairo steps and applications of each builtins, weighted by a vector: CairoResourceFeeWeights
Current weights:
#OnChainData = storage updates + L2 <> L1 messages + deployed contracts
- Storage updates
For each contract: the keys to update and the new values
- L2 <> L1 messages
L2 sender, L1 destination, payload
- Deployed contracts
Contract address, class hash, constructor arguments
Read 6 tweets
The size of #NLP models have increased enormously, growing to millions, or even billions, of parameters, along with a significant increase in the financial cost and carbon emissions. ASAPP Reducing the High Cos...
The cost associated with training large models limits the #AIresearch community's ability to innovate, because a research project often needs a lot of experimentation.
Consider training a top-performing #LanguageModel on the Billion Word benchmark. A single experiment would take 384 GPU days (6 days*64 V100 #GPUs, or as much as $36,000 using AWS on-demand instances)
Read 21 tweets
NEW #DTAT paper: What Were They Thinking? Pharmacologic priors implicit in a choice of 3+3 dose-escalation design arxiv.org/abs/2012.05301

“The unexamined trial is not worth conducting.”

— Socrates 1/

This latest #DTAT paper @arxiv sets out to reverse-engineer the unstated (∴ unexamined!) pharmacologic intuitions that underlie the #trialsafety claim implicit in the decision to conduct a dose-escalation trial. 2/ Image
Prior elicitation from doctors has never been easy, especially about #pharmacology or the future of the #Daleks: 3/

Read 39 tweets
Daily Bookmarks to GAVNet 08/27/2020 greeneracresvaluenetwork.wordpress.com/2020/08/27/dai…
Platforms vs Verticals and the Next Great Unbundling - Andreessen Horowitz

a16z.com/2019/09/11/pla…

#platforms #verticals
A strange new way to save lions and leopards—paint eyes on the butts of cows

anthropocenemagazine.org/2020/08/a-stra…

#lions #conflict
Read 6 tweets
Daily Bookmarks to GAVNet 08/24/2020 greeneracresvaluenetwork.wordpress.com/2020/08/24/dai…
Phenomenal World | Logistics, Labor, and State Power

phenomenalworld.org/interviews/lal…

#labor #state #logistics #power
How Does Commodity Trading Work? Not Well Enough for Some European Banks - Bloomberg

bloomberg.com/news/articles/…

#symbiotic #european #trading #memetic #banks
Read 7 tweets
Daily Bookmarks to GAVNet 07/31/2020 greeneracresvaluenetwork.wordpress.com/2020/07/31/dai…
Conserving Africa’s wildlife and wildlands through the COVID-19 crisis and beyond | Nature Ecology & Evolution

nature.com/articles/s4155…

#coronavirus #ecology #evolution #nature #africa #crisis #wildlife
Hidden figures in ecology and evolution | Nature Ecology & Evolution

nature.com/articles/s4155…

#ecology #nature #evolution
Read 6 tweets
These are the #stories of #math that we celebrate, amplify and make central to the idea of #maths as a #objective #neutral #culture-free subject. #Math can do no #harm, and the #logical “pure math” thinking of #Mathematicians is worthy of emulation. But... 1/x
2/x ... "What about pure mathematics and mathematicians who merely prove theorems? Is there any #ethical component comparable to what you find in other fields of #science?" - R. Hersh in Experiencing #Mathematics
3/x One point of view: "there's no need for #mathematicians to have a code of #ethics, because what we do matters so little that we can do whatever we like."
Read 34 tweets
Drawing Innovation: @illumina Super-Resolution Optics

Recently, $ILMN unveiled the NextSeq 2000, a mid-throughput sequencer w/ a ~2X reduction in operating costs ($/GB).

We think the upgrades inside the NextSeq 2000 may soon enable a $100 clinical-depth human #genome.

But how?
The cost savings come from (1) a new upfront #chemistry and (2) super-resolution optics. We think @illumina will use real-world customer data on the NextSeq 2000 to optimize 1 and 2, then possibly upgrade the #NovaSeq line.

illumina.com/systems/sequen…
I'm going to briefly (!) illustrate how #sequencing (*by synthesis*) works, walk through a farming analogy, then dive into some technical details. (Sources at the end)

Below is a general overview of a whole genome sequencing workflow (sample in, data out).
Read 19 tweets
Today's Applied Complexity Symposium begins with a panel on Computation and Complex Economies, feat. Rob Axtell, Joshua Epstein, @C4COMPUTATION, Blake LeBaron, John Miller, & @MelMitchell1

(Follow this thread for live coverage all day...)

#complexsystems #economics
4 Principles of Collective #Computation:

1 Ground truth (objective reality)
2 Effective ground truth (what we agree on, whether or not it is accurate)
3 #Information can be collectively encoded in #networks
4 Outputs are a product of collective dynamics

@C4COMPUTATION right now
"I think #economics really needs to adopt a generative framework with agent-based models of cognitively plausible software people...[and] we don't choose to fear."

- SFI'S Joshua Epstein (@GeorgeMasonU) on modeling emotions at today's ACtioN symposium:
Santafe.edu/action
Read 44 tweets
"Information theory, predictability, and the emergence of complex life"

SFI Seminar by Luis F Seoane of @IFISC_mallorca, streaming now:
"The size of the #environment in some ways captures the #complexity...longer words take longer and longer to sample. This means in smaller environments, smaller words mean more."

- Luís Seoane of @IFISC_mallorca on #information right now:

"More-complex bit guessers are going to survive in a wider variety of environments. But a more-complex guesser is going to pay a higher metabolic cost."

- Luís Seoane of @IFISC_mallorca

Streaming now:

#computation #intelligence #evolution
Read 5 tweets
Modular, hierarchical, and information-processing:

Right now at SFI, biologists, physicists, neuroscientists, and computer scientists convene for a three-day workshop on Biological #Computation, organized by David Wolpert, @C4COMPUTATION, & @albert_kao.

santafe.edu/news-center/ne…
"A circular definition isn't worthless if the circle is big enough."

- @livcomp of @UNM at today's #biocomp workshop
"#Physics produces order through the minimization of #energy. Adaptive systems produce order through the addition of #information processing."

By why do biological systems compute, whereas classical physical systems apparently don't? @C4COMPUTATION on #function at #biocomp now:
Read 4 tweets

Related hashtags

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!