ardivekar

Turing, etc. And yes.

I'm one of the organizers. This is a new experiment we are trying. We're grateful to have major academic and industry AI labs from both the US and China interested and participating.

Registration is still open until 31 July 11:59PM AoE. If you are a lab building an AI Scientist or ML autoresearch system, registration is free upto one entry. More details here: https://ioai-official.org/ai-model-track/faq-section/

Just finished with Terry Pratchett's Night Watch. So good.

So basically you're saying, having a million dollars is better than not having a million dollars :P

At least upload a preprint to arXiv or something. What's the point of citing a paper that doesn't exist yet?

I think it's a problem of reputation, actually. Google and Apple keep stuff online because they promised they would, and there would be backlash if they turned it off all of a sudden. There's nothing stopping a random, anonymous "filer" (like miner) from stopping the upload stream.

To go even further, if you orchestrate several miners carefully (maybe computationally) they can together perhaps stop people from accessing certain files.

Optimised code is occasionally so cryptic that you'll just waste a few hours figuring out what it does, which basically nullifies the effect of faster compile time.

That's "in vogue", isn't it?

I think I can answer your question, because I asked an identical one on bitcoin.stackexchange.com about a year back.

The transactions are public because that's the only way account balances can be tracked in this system. There's no 'amount' associated with an account id at any point in time, as there is for a banking database. In Blockchain, all the money an account has to spend is the net of the transactions made to and from that account. In traditional banking, even if you deleted the transaction history, it would work, because the account balances would be correct. This incentives people to fudge the history and the final balance via fraudulent transactions. Blockchain prevents that, since it never stores the balance, and old transactions are set in stone.

As to your second point, yes, mapping the wealth is possible. But so is obscuring the account holder's ID. To get your money, you just need a private key, a little string of characters, to verify that account. Also, account creation is incredibly easy (literally just a function call returning a 256-bit hash). So, no one knows who is who on the network. Some transactions may go from a person to themselves, others may go from a person to another for services rendered. It's basically impossible to tell once the transaction volume grows large enough.

Actually, only the -coin currencies have no actual value aside from transactions, which as can be shown, is very easily duplicable. I would say that Ethereum is an application platform, and that does have value (see WordPress, AWS etc).

Ripple also has value for ForEx settlement, which currently can take on the order of days. Also, it can function as a universal liquidator.

The -coins though, are ephemeral. Only the underlying Blockchain tech is of any use.

AKA half the work done in Machine Learning nowadays.

No no, results stored on The Blockchain.

RESnet in 2016 achieved 97% accuracy on the ImageNet challenge with hundreds of classes. I think that's near human accuracy.

I found "An Introduction to Statistical Learning with applications in R" to be very useful.

Which college and which program, if you don't mind sharing?

Edit: for the Master's degree.

I've heard of showerbeer, but never showercoffee. This has completely changed my outlook on consumables taken while bathing.

Technically yes. Practically, no.

To be honest, for personal use they aren't too useful due to their volatility. However, systems like Ripple and Ethereum could have potentially very big impacts for banks and businesses. They replace a cumbersome lower layer for which settlement time is usually on the order of days

> gif.png

This made me chuckle.

This really is true. During an interview I gave in college, I was rejected because of 'culture fit', because I tried to explain my solution to the interviewer but he wasn't bothered to understand it. My friend (who later got hired) told me that he said that while I was technically capable, I had 'an attitude problem'. Not saying I wasn't at fault, but it does matter because interviewers are human too, and like working with people they enjoy spending time with.

Really, check out WPS Office on both Linux and Windows. It's a phenomenal MS-Office clone, 100x better than both OpenOffice and LibreOffice (because it actually acknowledges that MS-Office is a good piece of software).

Paul has an entire book of his essays by the same title, it's a great read all kinds of programmers. :)

Maybe a two-way 'add contact' feature would be useful, like a facebook friend request.

Doesn't AWS do this too?

Not to sound rude but genuinely curious, why are you asking on behalf of your brother?