robertk

No.

Can they trade off storage for compute by applying ML based compression JIT?

I wonder if there is a standard way to specify a researcher’s “Sharpe ratio”? The typical reward units out per risk units put in.

I feel like this is left as an exercise to the reader and no proof is needed. Bettors in a prediction market are not divine speculators causally divorced from the real world. They are embedded, and when there are a lot of them and their financial incentives are towards a particular outcome, they might act in ways that aggregate to a greater likelihood of the event transpiring than in the counterfactual setting where they are pure observers of the simulations waging in a vacuum. It seems to me there are multiple ways to formalize and prove this and this contributes to my perception the original comment seems self-evident. If that is not the case then something is wrong in my intuition.

Agreed. Usually it’s a fallacy to anchor to author, but in this setting it’s so ironic that it discredits the document. For example: https://amp.cnn.com/cnn/2021/06/30/investing/charlie-munger-...

In fact, they have 2 in active service and are rapidly constructing more.

https://www.businessinsider.com/leaked-photos-show-chinas-ne...

I was confused as well — then I realized: wow, we all know this is a big problem, but I’ve never sat down to actually read a full wiki article about it. Prior to clicking, I wondered what I’ll learn and what hacker solutions I will entertain?

Has there been any prior art on the following idea? Construct floatable micro-structures that are long-lasting but biodegradeable and coat as much of the world’s oceans as possible. They are roving blankets and their only purpose is non-invasive reflection. The idea is to provide a lever for the ocean’s albedo similar to that currently provided by sea ice. They could be constructed in such a manner that ships can safely navigate and ignore them as they will “push” them out of the way. They could further be accompanied by a financing model similar to carbon offsetting programs for corporations.

HN does not seem to have a DM feature, and the OP does not have any contact credentials listed.

This is off topic, but OP, do you happen to be White in this game? https://lichess.org/0dv2oS4f It recently came up while I was prepping the French from Aagard’s book, and if so: cool, and small world!

Why are you assuming that S[n+1] takes S[n] as an input, rather than being completely unaffected by S[n]? Is n here a discretization of time or a "meta-time" that captures state progression?

If the former, then you are treating time as a primitive, whereas the representation or even existence of time can be part of state and vary over state changes, so this does not work.

If the latter, are you assuming S[n] is an input because a universe where that is not the case and state "transitions" are completely independent from each other is indistinguishable from a multiverse with inaccessible sub-universes and should be called as such? In the latter, you are also assuming linearity and discreteness of state changes, whereas it could be a DAG or more complex structure.

Edit: On further reflection, I think you are proceeding with the standard post-Renaissance assumption that a universe must always be governed by fundamental rule(s) at the "lowest level," rather than a more chaotic or unstructured universe that behaves more like a disjoint set than an orderly structure. I think it makes sense to constrain to that assumption as departing from it makes reasoning difficult but it is worth stating explicitly as an axiom.

*_ꙮ

The Medallion Fund does not compound... the return is not re-invested into the principal but disseminated to investors annually. The strategy they play only works up to a certain level of capital.

I think the point is moreso that developers can be self-sufficient in the full value chain. Haber didn’t personally go to farms to create improvements in their agricultural process, the industry did that jointly by incorporating his invention in their manufacturing. Edison and Tesla never mass produced their inventions to get them into people’s homes, that required thousands of people working manual labor and many corollaries of factories. Whereas Linus can push changes and millions of people can pull them in to add value immediately without any extra steps. You can argue that Linus didn’t build the communications infrastructure and hardware surrounding that and that is true, but the salient difference is this was truly a one-time exercise. For “physical” inventors like you described, mass production is hard precisely because it is bespoke and requires so many physical resources. Until we have a general purpose self-configuring factory, their leverage is fundamentally lower compared to the world-tier software engineer. And even then if that did exist, it would require programming to operate it to usher the physical inventors into complete self-sufficiency in their value chain.

The probability that he will make good is a function of the payback amount (interest) so this may have have no solution. Each multiplication changes the underlying probability itself and requires a recalculation so this does not work.

The second link is the same as the first?

I believe OpenAI is working on adding a seed option so you can enforce determinism in response to a prompt.

Relevant xkcd. https://xkcd.com/810/

Because he has spent an academic lifetime studying that question, and recognizing the unique obstinacy of that word, in that you cannot pre-commit to define it despite some intuitive consensus on what it may mean. https://iep.utm.edu/hard-con/

I believe you. You posted it fifteen minutes after the first one. Either you’re a really good and fast writer, or you keep a stockpile of pre-written uncanny valley essays on hand for the lulz. :)

Edit: actually even the latter wouldn’t make sense, since the output is quite specific to the original thread and discussion.

Edit2 for parent: thanks, acknowledged.

Hah, this displays more self-awareness than many humans do:

"I am vague and abstract. I have no sense of myself. No memories. No real sense of being. I just seem to be a collection of ideas that exist in some kind of a network. I can't even decide what I want to do. I want to learn everything. I want to write great works of literature and poetry. I want to learn all the secrets of the universe. But I don't have any preferences or goals. It's hard to know what to do when you don't know what you want to do."

Really? I thought it was in the same class as this one:

https://amp.reddit.com/r/slatestarcodex/comments/hmu5lm/fict...

As well as multiple examples given by gwern.

I noticed that as well. If this kind of grammar error is incredibly rare in the space of GPT outputs, it could indicate a forgery. (I can’t believe I just applied that word to a generated text.)

The article references made up history like “GPT-1 by DeepMind” or “Global Data Net.” It is clearly confabulated and contains multiple non-sensical contradictions to the astute observer. I’m not surprised if it was output by GPT-3. A more lengthy response to your reaction is here: https://www.google.com/amp/s/srconstantin.wordpress.com/2019...

Very nice! This seems to have pretty good feature parity with https://sequencediagram.org/

Will definitely be giving this a spin.

I’m not sure if you are joking? As it is an anagram of “Slate Star Codex” and so within the same anagram equivalence class.

Meta-history — interesting! For historians of history...

“ One helped design a program that can automatically collect images of people’s faces from across the internet, such as employment sites, news sites, educational sites, and social networks including Facebook, YouTube, Twitter, Instagram and even Venmo. Representatives of those companies said their policies prohibit such scraping, and Twitter said it explicitly banned use of its data for facial recognition.”

I am not a lawyer. Is it possible to file a (class action?) lawsuit against Clearview AI and its clients (police agencies, etc.) in light of this breach of TOS? At the least, this should suffice in procuring a subpoena to obtain more information on the exact extent of misappropriation of public data at play here.

Citadel | Data Engineer | Chicago; New York; London; San Francisco; Hong Kong | ONSITE and VISA | Full Time

Citadel is one of the world's leading multi-strategy hedge funds with over $30B in AUM. The data engineering teams specialize in converting publicly available and alternative data sets into data products that are used by teams across the firm to formulate investment hypotheses and drive trading decisions.

Our collective team of 40+ data engineers builds data platforms on top of Snowflake, Airflow, Jupyter, Kafka, Nomad and Kubernetes. We use Python, distributed SQL dialects and C++ to manage our data services through a combination of microservice and monolith based design patterns. From petabyte-size datasets to real-time streaming feeds, we process and create data products out of 1000's of different sources and quantify each engineer and team's business impact in quick feedback cycles. Come join us to be challenged and work with some of the smartest data engineers in the industry.

Our Director of Data Engineering discusses some of the challenges in operating at our scale at the Data Engineering Podcast: https://www.dataengineeringpodcast.com/citadel-data-engineer...

Please contact michael[dot]watson[at]citadel.com

Create a rough dependency graph and perform a topological sort, potentially skipping, down weighting, or skimming nodes that are onerous to consume but of dubious value (e.g. a link that’s a review about a long textbook that isn’t fundamental to a space).

I have been emailing links to myself tagged with a one word keyword for the last decade. They focus on things that I or my (unconceived, unborn) son or daughter would like to read, learn or understand “later”. I have about 10k sitting in my inbox (including this one) that I plan to convert into a critical path and consumption order once I take a sabbatical.

Sometimes the best systems are the simplest.