bearzoo

somewhat related is the number theoretic transform

https://ieeexplore.ieee.org/abstract/document/1451721

think the author is talking about having to fix the extra vector in V to be zeros and making sure to not compute/apply gradients to it

I think you are going to confuse people by framing this as "time series" and then focussing on reinforcement learning. A lot/most of your framing is around RL.

from a first glance and a read through your roadmap, this does not feel suitable for people who know what they are doing with RL. It also does not feel suitable for people who don't know what they are doing with RL.

Eh I feel like may be worth cutting some slack here..

It sounds like you’d like the title to be prefaced with “one time that”.

I don’t think anyone reading this says, hey gans are done for. I doubt the authors think that and I (clearly) don’t get the impression that’s what they mean.

Even the first sentence of the abstract:

> We show that diffusion models can achieve image sample quality superior to the current state-of-the-art generative models

Has the word current in front of state of the art generative models.

I don't think people have given up on model based RL, it is just that describing a proper model is (like you are saying) very difficult.

in the case you haven't seen or read the following: https://bair.berkeley.edu/blog/2019/12/12/mbpo/

many scientific computing applications are considered to be bounded by io

There are a lot of different definitions of intrinsic value. But it often has nothing to do with how much someone will give you.

There is the asset, an estimation of what the asset is truly valued at (intrinsic), and then finally the market price. In the case of the painting, the intrinsic value might be a function of a study of trends in demands of paintings, the artist, etc. The market price may or may not correctly factor in that information to price the painting well.

The fact that insider information does exist should paint a clear picture of why someones calculation of intrinsic value might be different from the market price.

Lastly, if someone gave me Hertz stock, I'd sell it ASAP and invest the money in something else.

https://en.wikipedia.org/wiki/Intrinsic_value_(finance)

Do you have any evidence that they are lying about all of the use cases they list on that web site?

actually, I think this is exactly what misappropriation theory is supposed to account for. can you explain more around why it is well within their rights? I don't think it is.

Svms are, by default, linear models. The decision boundary in the Svm problem is linear and since it’s the max margin we may enjoy nice generalization properties (as you probably know).

You probably also know that decision tree boundaries are non Linear And piecewise. It’s not so straightforward to find splits on continuous features.

Ie If the data is linearly separable then why not. Even using hinge loss with nns is not uncommon.

You probably see gbms winning a lot of competitions compared to svms because a lot of competitions may have a lot of data and non linear decision boundaries. some problems don’t have these characteristics.

in NYC una is the best i've found. it easily competes with the best in naples (in my opinion)

Thanks, that is pretty impressive

Oh? I don't suppose that julia wraps lapack functions for no good reason. What makes you say this

The flaw, in my opinion, is not necessarily in the benchmark but in the claims and observations around the benchmark. For instance the blog flat out says:

> Nope, we work with float32.

It seems that is not true. The blog should make clear that it isn't straightforward or possible to get cupy to do this.

This is a bizarre analysis in my opinion. It’s not all about the mortality rate, it’s also about spending in general.

Have you experimented with replacing the averaging operation on the vectors with a recurrent network such as an LSTM. This way you can not ignore the temporal nature of the feedback (I have had success improving metrics doing this on implicit streaming video feedback).

I would email greg brockman. I pestered him for advice one time and the amount of time he took to respond (and the amount of times he responded) was well beyond my expectation.

A "solved" problem? I think that is a ridiculous thing to say. Just because people know how to parse json does not mean there aren't solutions that are more performant or useable. Shit maybe they just wanted to parse and collect frequency counts of keys or something all in the same pass? There are reasons to do so.

thanks for all the great responses!

With spearmint I had the ability to modify the parameters of the mcmc sampling (e.g. burn in iterations). Will sigopt expose parameters for those of us who want to manipulate them? Will there be options to use different types of function estimators to estimate the mapping between hyper params and performance (i.e. what if I would like to use a neural network or a decision tree instead of gaussian processes)?

I say these things because as someone who is active in machine learning - I often want to optimize hyper parameters. The type of people that are serious about optimizing hyper parameters (i.e. people who may not like to use grid or random searches) for a model are usually some what technical. Your product seems to be catered to those who may not be too technical (very simple interface, etc). How will you balance what you expose in the future without giving away too much of your underlying algorithms?

Well - just my two cents. The title feels inaccurate. You all are tuning hyper parameters with respect to the performance of the classification task. The bayesian optimization is really to optimize the unsupervised -> supervised pipeline. I was expecting some bayesian optimization of strictly unsupervised representation learning (ex. we have an autoencoder and use some bayesian optimization to tune hyper parameters in order to minimize a reconstruction error). This is really just supervised learning with even less supervision (which is quite typical).

For work a month ago I implemented stochastic gradient descent in some arbitrary directed graph of operations (I applied a skill I learned from neural networks in a system that had nothing to do with neural networks) which required me to know how to take derivatives of a loss function with respect to parameters. I did all of the math on paper, and then coded it and it worked (to be frank this surprised me). It often may be hard to see the immediate gains from understanding things in a more technical, low level fashion - but I think in the long run they give you a great deal of power.

I did not claim anything about the underlying data.. The 2 dimensional embeddings were forced into the unit circle because the 'perplexity' hyper parameter for the t-sne was set too high.

From the guy who helped make t-sne:

When I run t-SNE, I get a strange ‘ball’ with uniformly distributed points?

This usually indicates you set your perplexity way too high. All points now want to be equidistant. The result you got is the closest you can get to equidistant points as is possible in two dimensions. If lowering the perplexity doesn’t help, you might have run into the problem described in the next question. Similar effects may also occur when you use highly non-metric similarities as input.

i once read that high perplexity can generate embeddings that are very tightly bound on a unit circle...not sure if this is what is going on

I REALLY REALLY want someone to do this with a huge amount of google image thumbnails so that the google crawlers just start hitting google servers. Would it be considered malicious to do such a thing?

Contrary to what many people will tell you - a degree is just ONE of the things you go to college for. Go to college. Your at a great age, and will have experiences you will not be able to have at any other time in your life. It is worth it, it will help you grow. You can keep a client or two on the side to pay bills, etc..

I am afraid this author has no idea what he is doing - and is loosely throwing around terms he does not understand. What the hell was his normalization procedure. Dangerous to readers who do not know a lot and will get confused while reading.