ghgr

I'm curious how efficient is RL for trading. Given that RL is extremely data-hungry I'd say that this approach is likely to result in an overfit model. Does anybody have positive experiences with this?

Minor correction: it's not your (i.e. the camera's) XYZ position what you input, but the position of the point whose RGBA you're trying to render.

I often come to HN for the discussions, not for the links themselves, and unfortunately that's something missing in "New". It happens quite often that the submission itself is not terribly interesting (e.g. a blog post ranting about some database engine), but nonetheless it spawns a fascinating discussion, filled with first-hand anecdotes and other insights.

So when I browser "New" I usually upvote submissions that have potential to create an interesting discussion (so I upvote what I believe _other_ people will find interesting, in a Keynesian beauty contest way).

The papers states that human understandable and computationally efficient are often very opposed goals.

It reminds me of the short story "Manna – Two Views of Humanity’s Future" [1], which describes the steady establishment of AI process optimizers as the managers of human employees, who increasingly became a replaceable commodity. Fascinating read.

[1] https://marshallbrain.com/manna1

I switched from Borg to Restic because it nicely integrates with cloud backup services. I've been using restic+backblaze happily for the last 2 years.

While I appreciate its simplicity, private data on the internet is messier than it sounds, and almost impossible _not_ to collect it (server logs store the ip by default for example). In particular, 13bananas.com has outbound connections to stats.wp.com (prompty blocked by uBlock Origin) and to bandcamp.com. Also, uBlock Origin seems to log access to Google Analytics through a bandcamp script.

Sonnet.io seems to be less chatty, except for the requests to Heroku.

Congrats on you game! It looks really nice, and I wish you the best of lucks!

> So I have a dedicated hard-drive that I boot into that is strictly for coding. Another is strictly for news & social media. This way if I'm coding, I am less likely to slide into Hackernews/Reddit/Twitter/Whatever. Friction is good and more people need to leverage it.

But then you're also less likely to get back to work once you are in social media, aren't you?

> in python `for` loop are equally unbounded as `while` ones. Yes, most of the time you are looping over a (finite) list or a (finite) dict, but the concept underlying the `for` loop is the iterator, not the list. And iterators can easily be infinite.

Not only iterators, but also plain lists. Example:

  a=[1]
  for i in a:
    a.append("the ride never ends")
    print(i)

The you-tldr seems pretty useful, although admittedly hard to monetize.

Do these alternative Smartphones are able to seamlessly run native Android/iOS apps in containers? It'd be great if they could "embrace" such walled gardens, and possibly "extend" some of their functionality :-)

Maybe the book "Ratio: The Simple Codes Behind the Craft of Everyday Cooking" [1] is what you are looking for:

[1] https://www.goodreads.com/book/show/3931154-ratio

> When I containerized MongoDB, Docker helpfully inserted an allow rule into iptables, opening up MongoDB to the world.

Yeah I had the same problem. If you are not using orchestration engines like Swarm of Kubernetes you can just avoid it by explicitly binding the port to your local interface (i.e. -p 127.0.0.1:27017:27017 instead of just -p 27017:27017)

Regarding this topic I came across the subject of Open-endedness [1] and the fascinating works of Jeff Clune, Ken Stanley and their colleagues (the two former are currently working at OpenAI).

EDIT: I added this paper [2] by Jeff Clune, a nice introduction of Open-endedness and their potential for reaching general AI.

[1] https://www.oreilly.com/radar/open-endedness-the-last-grand-...

[2] https://arxiv.org/abs/1905.10985

It took me a while to find out the description of this proof of space-time (and check whether it's satire or not), so to save you time here's the gist of it:

From the whitepaper:

Loosely speaking, the PoST consists of two phases: an initialization phase (executed once), in which miners “commit” to the data that fills the space S , and an execution phase (executed repeatedly), in which miners prove that they are still storing the data. The time component of the spacetime resource is the elapsed time between successive proofs—if the interval between initialization (or the previous execution phase) and the latest execution phase is T, this proves the miner expended S · T spacetime.

https://drive.google.com/file/d/18I9GPebWqgpvusI1kMnAB9nayBb...

That's an insightful experience, thank you for sharing. To give a counter-counter point, in my experience working with LIDAR (terrestrial in my case, not airborne) there's a wealth of tools that claim to process large amounts of unstructured pointcloud data and make all kinds of fancy visualizations and 3d models (Pix4D comes to my mind). The "catch" was often to find high quality data to feed these tools (other than the sample data provided by them). I'm convinced that having high-quality proprietary data is the moat the investors are looking for, not the algorithms. After all, universities are cranking improved algorithms all the time for free.

ESG probably stands for Environmental, Social, and Corporate Governance: https://en.wikipedia.org/wiki/Environmental,_social_and_corp...

For speech-to-text extraction you can try Silero [1].

Free software (AGPL-3.0 License), fast, highly accurate and extremely simple to deploy (I have no affiliation with them).

[1] https://github.com/snakers4/silero-models

Agreed. And not only that, in my case at least the Signal desktop app is randomly missing messages (and 100% of the images). I tried reinstalling it multiple times to no avail (Ubuntu). I'd really love to see more 3rd party alternatives.

An important point that I didn't see mentioned: Signal implements the "Sealed Sender" feature in order to encrypt metadata (as much as possible).

https://thehackernews.com/2018/10/signal-secure-messaging-me...

That's in my opinion the USP against other (actually or allegedly) E2E platforms like WhatsApp or Telegram.

I wasn't aware of that paper, but it seems we came to a similar conclusion. From the paper:

> Across all sports, scoring tempo - when scoring events occur - is remarkably well-described by a Poisson process, in which scoring events occur independently with a sport-specific rate at each second on the game clock.

So, yes, basically that was my concept. Estimate its parameters and then you'll have the "real" odds. Any difference with the market ones is a potential of profit (considering the house fees).

I couldn't backtest it since I couldn't find free historical data for all types of markets (then again, maybe I didn't look hard enough). I forward tested it, though, and it seemed to work.

My code is here [1], which also links to a blog post describing the method. Feel free to take a look and contact me to discuss these ideas.

Also, if you're interested in mathematically modelling probabilities, we are looking for team members to develop a product to estimate the risk of default in loans. After all, a loan is another kind of bet, that takes place in a slightly harder to model environment.

[1] https://github.com/ghgr/BetFair_Arbitrer

Thanks for sharing (I recommend taking a look at his GitHub repo [2], where he nicely described the game and his approach).

Some time ago I had a similar project with BetFair soccer matches. I assumed that the number of goals from each team is given by a Poisson process. Then, I used the largest market (Win/Lose/Draw) to estimate this parameter. With this information I could estimate the "fair" odds of more exotics bets (like e.g. team A will score 3 goals more than team B), which were often mispriced, and then use the Kelly criterion to estimate how much to bet.

All in all a nice project, but... at the very end -just before deploying the bot- I realised that the BetFair Exchange is not open to Germans, so I wrote a blog post and open sourced the code. Still, a nice learning opportunity :-)

I can attest that -even with kids- this is almost set-and-forget.

  sudo docker run -d --restart=always -v <LOCAL_PATH_TO_NEXCLOUD_DATA>:/var/www/html --name nextcloud --network=host nextcloud

If you want to get fancy with a MySQL database just make:

  sudo docker run -d --restart=always -v <LOCAL_PATH_TO_MYSQL_DATA>:/var/lib/mysql --name=mysql --network=host -e MYSQL_ROOT_PASSWORD=hunter2 mysql

> [..] even Google's own forum (https://groups.google.com/g/android-developers) appears to be only a honeypot to train their spam protection :)

I thought you were exaggerating, but dear Lord, it's really full of spam (yes, Google themselves say that "Posts about job openings or from recruiters are considered spam."). Just a single legitimate question, which seems more general support than a technical question by a developer.

It's a modern misrepresentation. In fact, Stoicism encourages to help your neighbor.

From the Meditations by Marcus Aurelius:

> [...] Do not be carried along inconsiderately by the appearance of things, but give help to all according to thy ability and their fitness [...] (Book V)

> [...] Reverence the gods, and help men. Short is life. There is only one fruit of this terrene life, a pious disposition and social acts. [...] ( Book VI )

You can read the full ebook here [1] (I'd link to project Gutenberg but unfortunately they're still blocking German readers). Alas.

[1] https://standardebooks.org/ebooks/marcus-aurelius/meditation...

That's odd, I've been using i3 for years now and modals always open as a floating window. Even GIMP (and its many windows) works fine. I don't remember tweaking anything in particular to make it work.

One practical advice is to install uBlock Origin + Privacy Badger and then just accept whatever they ask you to. Otherwise you'll always wonder what "essential" really means to them.

Edit: uBlock Origin, not uBlock

Well at least Decentraleyes states in their page that "Complements regular content blockers" [1]

[1] https://addons.mozilla.org/en-US/firefox/addon/decentraleyes...

I wonder how browser based 3D rooms work in real life, so I just created a room and share the link with you. Enter and say hello, no registration needed.

Disclaimer: Frame VR belongs to virBELA, a company that seems to develop this kind of solutions. Not affiliated with them in any way.

Same. I'd find such a "feature" extremely dangerous.