Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Federated Learning

federated.withgoogle.com

Post reply on HN
Print view View original

51–60 of 77 posts

Re: Federated Learning

#51

Post by cromwellian »

krick wrote: post #48
cromwellian wrote: post #39

Earlier quoted context omitted.

This is not the way it works at all (see Secure Aggregation). There are a number of techniques out there that permit privacy safe services and learning, like Differential Privacy, Federated Learning, and there's even Deep Neural Nets using homomorphic computing (e.g. https://arxiv.org/abs/1711.05189 ) How about we lay off the conspiracy theories every time any new paper is published.

"Secure aggregation" doesn't mean a thing, because I would argue, as I did in another comment already, that "you" are not "your device".

You can put as many words as you want in quotes, but if you can point out a specific flaw in the secure aggregation paper, either in the algorithm description, or the math, it would be taken a lot more seriously.

It's like your responding to the Bitcoin paper by saying "proof of work" doesn't mean a thing because "hashing" isn't "money".

There are papers published, critique those specifically, instead of relying on a handwavy respond to a comic.

https://ai.google/research/pubs/pub47246

Re: Federated Learning

#52

Post by pm90 »

krick wrote: post #46
capitalsigma wrote: post #35

Earlier quoted context omitted.

It's legitimate to call the aggregate data that the central server has "non-private." For example, it's ok to publish how many thousands of cars drive on a particular highway -- that's public data. It's not ok (broadly) to publish what route John Doe drives to work everyday -- that's private data.

Absolutely not. Actually, I kind of reflected it in my concerns: it makes it easier for somebody who wants to push such a polemic (as yourself) to do so, but by no means it is universally true. I mean, it might be, if everything that your model learns is the number of cars on a highway. But there's absolutely no reason to assume it is. It might as well learn anything else about your private life. You might be tempted…

What if Google understood the more meaningful you but couldn't track it down exactly to the real you? Or is that impossible to achieve technically?

Re: Federated Learning

#53

Post by dmix »

pm90 wrote: post #42
grantlmiller wrote: post #6

First, I've loved that Google open sourced Tensor Flow Federated as a way to encourage the rest of the world to adopt this method of decentralized machine learning. Second, I was a bit disheartened that this concept had to be explained with a comic strip to make it accessible because I hoped the benefits were clear to everyone. Third, I read the comic strip, learned new things (secure aggregation protocol, wtf, amazi…

Popular culture is a tool for the education of the masses. Even for people who may be technically inclined, its not always evident what certain technologies really do. I am a software engineer but mostly work on DevOps-y stuff. This was a very accessible, low-investment way for me to understand exactly what "Federated Learning" really meant. Some of the best teachers at Univ had a way of explaining things in simple t…

Universities have a captive audience, they can take the time to walk you through incrementally.

Most websites and online communication don't have that luxury. It's interesting how the comic works well in these situations, while still pushing a long-read format. Google did the long-form comic thing with Chrome too and I remember reading it page-to-page back then.

But at the same time, is it a good idea as your primary website homepage as it is here? Which would be unusual if there was anything more to it like documentation, code, etc. Right now this website is clearly in an educate-the-public mode only which is how they can get away with this being the primary content.

Re: Federated Learning

#54

Post by krick »

cromwellian wrote: post #51
krick wrote: post #48

Earlier quoted context omitted.

"Secure aggregation" doesn't mean a thing, because I would argue, as I did in another comment already, that "you" are not "your device".

You can put as many words as you want in quotes, but if you can point out a specific flaw in the secure aggregation paper, either in the algorithm description, or the math, it would be taken a lot more seriously. It's like your responding to the Bitcoin paper by saying "proof of work" doesn't mean a thing because "hashing" isn't "money". There are papers published, critique those specifically, instead of relying on a…

It is somehow understandable for a Google employee to take such a defensive position, but you either didn't read or didn't try to understand what I'm saying. I don't believe there is anything wrong with federated learning or that there are any problems with "secure aggregation" you are so fondly referring to. It's just that they don't mean shit, because, as I said several times already, you don't need to know that some data comes from my device, to learn something sensitive about me. It depends entirely on the application, no crypto has to be broken in order for that to happen.

Re: Federated Learning

#55

Post by strin »

ma2rten wrote: post #47
walterbell wrote: post #25

Google mentioned at I/O that speech recognition will soon (this summer?) be performed locally on Android devices, with no voice data being sent to Google, because they have been able to reduce the size of the model dramatically. Is that related to federated learning? Paper: https://arxiv.org/abs/1811.06621

No, federated learning is about training on the device not running prediction on the device. Training a speech model on the device would be hard, because there is no labeled data. We don't know what the user said.

I can imagine the world relying more and more on unsupervised pre-training approaches, such as BERT and GPT-2. Then we’ll just need a few labeled data to generalize.

Re: Federated Learning

#56

Post by defen »

cavisne wrote: post #18
ahelwer wrote: post #2

All right, I'm cynical as all heck about ad companies and privacy, but this has me optimistic. Somebody disillusion me, why shouldn't I be optimistic?

Well the cynical view would be. 1) This still lets you have personalized models, just trained on more than 1 user, thats fine at google's scale anyway 2) Their competitors (FB, AMZN) dont have the edge compute (Android) to do this, and to a lesser degree don't have the ML stack (however Android implements this at the API level will be very Tensorflow focused) 3) Now google can push for privacy regulations that preven…

> Their competitors (FB, AMZN) dont have the edge compute (Android) to do this

It's not Android scale but Amazon has sold 100 million Alexas.

Re: Federated Learning

#57

Post by krick »

pm90 wrote: post #52
krick wrote: post #46

Earlier quoted context omitted.

Absolutely not. Actually, I kind of reflected it in my concerns: it makes it easier for somebody who wants to push such a polemic (as yourself) to do so, but by no means it is universally true. I mean, it might be, if everything that your model learns is the number of cars on a highway. But there's absolutely no reason to assume it is. It might as well learn anything else about your private life. You might be tempted…

What if Google understood the more meaningful you but couldn't track it down exactly to the real you? Or is that impossible to achieve technically?

Does it matter? I'm saying that what you mean by "the real you" may not really be the most useful definition. If there are 10 people on the planet that would behave exactly the same way when put in the same conditions, I don't really need to make a distinction between any of you. Each one is as good as the other.

I don't even feel necessarily comfortable with the completely anonymous usages of ML on my data. Like the mentioned "next word prediction". Language models we've seen by now don't really understand anything about the text, and surely nothing about who you are. Yet they are uncannily good in "understanding" the context somehow. It really doesn't know anything about the world in the strict sense of the word, but given your sentence starts with "Tensorflow" it is still able to understand that something about "neural networks" and "machine learning" would be a good way to continue.

So if it learns on some very unique stories on a very unique subject you were telling someone in the WhatsApp, a model, trained on this data, actually might tell someone else the story vaguely resembling what you just said given the right context. Even though it didn't try to learn anything specifically about you, or even gather any data from you in a non-anonymous way.

Of course, I don't mean to say this is actually likely to happen with how it's likely to be used, I'm just saying that to illustrate the possibility.

Re: Federated Learning

#58

Post by CyanTas »

arthurcolle wrote: post #41
ddtaylor wrote: post #40

Earlier quoted context omitted.

Homomorphic Encryption

Does this actually work today? I was tangentially involved in some 'zero protocol'/zcash-related projects a few years back and the lack of ability to communicate and transfer information while being able to perform computation on it was a major drawback to most of the interesting ideas in the space. Are they actually using this in this intended federated learning plan? If so that's a truly major innovation.

Fully homomorphic encryption, in which you can do arbitrary computation on encrypted data, is still quite slow. But partially homomorphic decryption, in which you can add encrypted values together but not multiply (or vice versa), is quite efficient. And since the secure aggregation protocol only needs to add together encrypted values to get an average, it only needs partially homomorphic encryption properties.

Re: Federated Learning

#59

Post by im3w1l »

My gut feeling tells me not to believe their promises that it's impossible to deduce the data from the model updates. That there should be attacks.

My stylistic criticism is that they portray white men in a demeaning way that they would never dare do to any other group.

edited to make a weaker claim

Re: Federated Learning

#60

Post by ximeng »

Linked paper on using this for Google Keyboard (https://arxiv.org/pdf/1903.10635.pdf) highlights that there are nevertheless still privacy issues with this approach:

While Federated Learning removes the need to upload raw user material — here OOV words — to the server, the privacy risk of unintended memorization still exists (as demonstrated in (Carlini et al., 2018)). Such risk can be mitigated, usually with some accuracy cost, using techniques including differential privacy (McMahan et al., 2018). Exploring these trade-offs is beyond the scope of this paper.

Post reply on HN