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

Differential Transformer

arxiv.org

Post reply on HN
Print view View original

131–140 of 188 posts

Re: Differential Transformer

#131

Post by testdfkjahdfh »

magicalhippo wrote: post #79
iandanforth wrote: post #58

The key bit I didn't understand at first was what happens if the two groups of attention learn the same thing; because their attention masks are subtracted from one another if they both output similar values the attention across the board will drop to zero and this will lead to high loss. So the only way to reduce loss is if they learn to attend to different things. One of the simplest strategies they could learn (an…

An interesting aspect is that they don't do a plain subtraction, but rather subtract a portion of the second softmax. This makes sense, if one considers that the two copies are identical then the softmax outputs would be identical and the difference is zero everywhere. However, by subtracting a scaled copy, the normalization of the difference seems to really boost the signal value(s) over the "noise", making the sign…

if two attentions A, B are identical, would (A - lambda * B) be just (1-lambda) * A, how does it "boost the signal value(s) over the "noise""?

Re: Differential Transformer

#132

Post by patcon »

iandanforth wrote: post #58

The key bit I didn't understand at first was what happens if the two groups of attention learn the same thing; because their attention masks are subtracted from one another if they both output similar values the attention across the board will drop to zero and this will lead to high loss. So the only way to reduce loss is if they learn to attend to different things. One of the simplest strategies they could learn (an…

> what happens if the two groups of attention learn the same thing

I wonder if there's a metaphor here for our own experience and utility in "surprise".

Like if one attention head is surprised by what another learns, up-weight it. But if they both find the same, assume it's not very surprising and down-weight it.

Admittedly, "surprise" is something that has a big section of my knowledgebase[1][2][3] (both as a subjective feeling and adaptive function of our minds, one of the most complex adaptive system we know of)

[1] https://plus.maths.org/content/information-surprise

[2] https://blakeelias.name/papers/Multi-Agent-Cooperation-Intri...

[3] https://complexity.simplecast.com/episodes/81/transcript

Re: Differential Transformer

#133

Post by mistercheph »

msoad wrote: post #7

Like most things in this new world of Machine Learning, I'm really confused why this works? The analogy to noise-cancelling headphones is helpful but in that case we clearly know which is signal and which is noise. Here, if we knew why would we even bother to the noise-cancelling work?

I think common mode filtering in balanced audio cables is a much better analogy than noise canceling headphones (and where this paper gets its name from I assume), you don't know what the noise is ahead of time, but if you take two samples with one positive and one negative, noise displaces both absolutely, which you can take advantage of to denoise the signal (find the differential mode).

For example, if you are trying to send a +1V signal on one wire, and a -1V signal on the other and a +0.5V noise exists, one wire will have +1.5V and the other will have -0.5V,

Take the difference and divide by 2:

(+1.5V - -0.5V) / 2 = +1V or, if your setup is different (-0.5V - +1.5V) / 2 = -1V

Re: Differential Transformer

#134

Post by ExxKA »

d3m0t3p wrote: post #19
ExxKA wrote: post #3

Very interesting. Currently working on timeseries with Transformers. Let me know if anyone else out there is also reading it from that context.

Really cool, I'm a CS majoring in AI, but I'm also interested in that domain, would you have any recommendation to get started ?

Get a lot of data, and just dig in :) No better way to learn.

Re: Differential Transformer

#135

Post by jszymborski »

x1000 wrote: post #117
blackbear_ wrote: post #43

Earlier quoted context omitted.

With a single softmax you cannot predict exactly 0, but only very small numbers. When you have a large number of values to add up, this "poisons" the output with a lot of irrelevant stuff (the noise mentioned in the paper). To make things worse, low attention values will have very low gradient, thus needing a lot of weight updates to undo that kind of mistakes. On the other hand, subtracting the output of two softmax…

Could you help explain how we would achieve an attention score of exactly 0, in practice? Here’s my take: If we’re subtracting one attention matrix from another, we’d end up with attention scores between -1 and 1, with a probability of effectively 0 for any single entry to exactly equal 0. What’s more, the learnable parameter \lambda allows for negative values. This would allow the model to learn to actually add the…

Your comment brings up two interesting variants that could be interesting if your goal is to increase the sparsity of the attention:

- Rectify the difference of the softmaxes. (min(0, s(A1) - lambda s(A2)))

- Apply the Heaviside function to the second softmax. (softmax(A1) - lambda H(s(A1) - lambda s(A2))

The second one being a bit more drastic and maybe harder to train.

Re: Differential Transformer

#137

Post by Jerrrrrrry »

watsonmusic wrote: post #119
Imnimo wrote: post #108

I feel like I'm missing a key insight here. I understand the problem that regular softmax attention struggles to approach assigning zero attention to irrelevant stuff. And I get that having this subtraction formula makes it possible to assign exactly (or near) zero attention weight without having crazy outlier activations. But it seems like it also makes it very easy to have negative attention weight (which is equiva…

negative values can enhance the expressibility

doubt is the seed of reason

Re: Differential Transformer

#138

Post by chessgecko »

msoad wrote: post #7

Like most things in this new world of Machine Learning, I'm really confused why this works? The analogy to noise-cancelling headphones is helpful but in that case we clearly know which is signal and which is noise. Here, if we knew why would we even bother to the noise-cancelling work?

My hypothesis for why this works that it mitigates the downsides of rope

to eli5:

rope is the modern strategy used to give information to the model about how far a query and a key are apart when doing attention. It's the best strategy we have now, but has a major downside, where it makes some connections between tokens that are far apart much stronger than you would like them to be. Xpos (https://arxiv.org/pdf/2212.10554) is another paper by microsoft tackling issues with rope and you can see figure 1 on page 4 to get a visual interpretation of the sinusoidal attention strength (you would like it to be smooth).

I think a big reason differential transformers is working so well, especially on long sequence stuff, because when both q1 and q2 don't match a token, the rope relative strength will still have the same value and the noise will cancel out. Leaving intended matches, but at the cost of somewhat dampening the original value rope brought.

Just a hypothesis though. It would be easy to test by running this experiment against a baseline where both use alibi attention (https://arxiv.org/pdf/2108.12409) which has a different set of tradeoffs this wouldn't mitigate, but still a really interesting result.

Re: Differential Transformer

#139

Post by Grosvenor »

Imnimo wrote: post #108

I feel like I'm missing a key insight here. I understand the problem that regular softmax attention struggles to approach assigning zero attention to irrelevant stuff. And I get that having this subtraction formula makes it possible to assign exactly (or near) zero attention weight without having crazy outlier activations. But it seems like it also makes it very easy to have negative attention weight (which is equiva…

Regular softmax (and attention) has an error in it.

softmax should be exp()/1+∑exp()

Notice the 1 added to the denominator.

The difference is at the negative limit, softmax can be 0, instead of some epsilon. The same could be done by adding an extra zero value in x.

Downside is, you have to retrain your model from scratch to fix this.

Re: Differential Transformer

#140

Post by vessenes »

aDyslecticCrow wrote: post #78

Very clever. I like this kind of nitty-gritty detail work, and the change is small enough to be adapted easily by others. Bravo! I'm a little concerned about the last sentence of the section introduction of "2 Differential Transformer". It mentions using improvements from previous papers, but in the grammatical context, it's unclear if this improvement is added to both the normal transformer and their diff transforme…

Yes. This looks really, really good to me. Cross the board improvements in training time, perplexity improvements per both token trained and per model size. I'm reminded of MoE architectures, in that world we're choosing an optimal small model to process part or all of the inference job; I wonder if MoE got some of the same benefits from forcing the Transformer to distinguish between alternate possibilities.

In any event, I'd imagine that this will get widely adopted if the numbers hold up; like I said, this seems to be basically no downside, and should be easy to replicate.

Post reply on HN