Ilya Shumailov

a person the same two sentences and then like the next person says the same two sentences and it usually gets like more and more garbled as it goes down the line i think this is a comparison kind of kind of works yes so this is the first thing it's the improbable events and then the second thing that happens is your models are going to produce errors so misunderstandings of the underlying phenomenon right and as a result

a person the same two sentences and then like the next person says the same two sentences and it usually gets like more and more garbled as it goes down the line i think this is a comparison kind of kind of works yes so this is the first thing it's the improbable events and then the second thing that happens is your models are going to produce errors so misunderstandings of the underlying phenomenon right and as a result

a person the same two sentences and then like the next person says the same two sentences and it usually gets like more and more garbled as it goes down the line i think this is a comparison kind of kind of works yes so this is the first thing it's the improbable events and then the second thing that happens is your models are going to produce errors so misunderstandings of the underlying phenomenon right and as a result

what you will see is that those errors start propagating as well. And they are relatively correlated. If all of your models are using the same architecture, then it's likely to be correlatedly wrong in the same kinds of way. So whenever it sees errors, it may amplify the same errors that it's observing.

what you will see is that those errors start propagating as well. And they are relatively correlated. If all of your models are using the same architecture, then it's likely to be correlatedly wrong in the same kinds of way. So whenever it sees errors, it may amplify the same errors that it's observing.

what you will see is that those errors start propagating as well. And they are relatively correlated. If all of your models are using the same architecture, then it's likely to be correlatedly wrong in the same kinds of way. So whenever it sees errors, it may amplify the same errors that it's observing.

Yeah, so approximations of approximations of approximations end up being very imprecise. As long as you can bound the errors of your approximations, it's okay, I guess. But yeah, in practice, because machine learning is very empiric, quite often we can't.

Yeah, so approximations of approximations of approximations end up being very imprecise. As long as you can bound the errors of your approximations, it's okay, I guess. But yeah, in practice, because machine learning is very empiric, quite often we can't.

Yeah, so approximations of approximations of approximations end up being very imprecise. As long as you can bound the errors of your approximations, it's okay, I guess. But yeah, in practice, because machine learning is very empiric, quite often we can't.

Yeah. So an important thing to say here is that... The settings we talk about here are relatively hypothetical in a sense that we are not in the world in which, you know, today we can build a model and tomorrow they disappear. That is not going to happen. We already have very good models and the way forward is having even better models. And there is no doubts about this.

Yeah. So an important thing to say here is that... The settings we talk about here are relatively hypothetical in a sense that we are not in the world in which, you know, today we can build a model and tomorrow they disappear. That is not going to happen. We already have very good models and the way forward is having even better models. And there is no doubts about this.

Yeah. So an important thing to say here is that... The settings we talk about here are relatively hypothetical in a sense that we are not in the world in which, you know, today we can build a model and tomorrow they disappear. That is not going to happen. We already have very good models and the way forward is having even better models. And there is no doubts about this.

I mean, there are many different solutions. You'll find a lot of different papers that are exploring what are the most effective mitigations. And it's mostly data filtering of different kinds. And basically making sure that the data that ends up being ingested by the models is representative of the underlying data distribution.

I mean, there are many different solutions. You'll find a lot of different papers that are exploring what are the most effective mitigations. And it's mostly data filtering of different kinds. And basically making sure that the data that ends up being ingested by the models is representative of the underlying data distribution.

I mean, there are many different solutions. You'll find a lot of different papers that are exploring what are the most effective mitigations. And it's mostly data filtering of different kinds. And basically making sure that the data that ends up being ingested by the models is representative of the underlying data distribution.

And whenever we hit this limit and we see that our model diverges into some sort of a training direction,

And whenever we hit this limit and we see that our model diverges into some sort of a training direction,

And whenever we hit this limit and we see that our model diverges into some sort of a training direction,

trajectory that is making the model worse i promise you people will stop training of the models retract back a couple of steps maybe add additional data of certain kind and keep on training right because we can always go back to previous models nothing stopping us and then we can always spend more effort getting high quality data paying more people to create high quality data

trajectory that is making the model worse i promise you people will stop training of the models retract back a couple of steps maybe add additional data of certain kind and keep on training right because we can always go back to previous models nothing stopping us and then we can always spend more effort getting high quality data paying more people to create high quality data