[HN Gopher] Deep Learning Is Not So Mysterious or Different
___________________________________________________________________
Deep Learning Is Not So Mysterious or Different
Author : wuubuu
Score : 233 points
Date : 2025-03-17 16:47 UTC (6 hours ago)
(HTM) web link (arxiv.org)
(TXT) w3m dump (arxiv.org)
| cgdl wrote:
| Agreed, but PAC-Bayes or other descendants of VC theory is
| probably not the best explanation. The notion of algorithmic
| stability provides a (much) more compelling explanation. See [1]
| (particularly Sections 11 and 12)
|
| [1] https://arxiv.org/abs/2203.10036
| bigfatfrock wrote:
| I'm a huge fan of HN just for replies such as this that smash
| the OP's post/product with something better. It's like at least
| half the reason I stick around here.
|
| Thanks for the great read.
| hlynurd wrote:
| >smash with something better
|
| Not a fan of the aggressive rhetoric here...
| superidiot1932 wrote:
| I too felt threatened
| pessimizer wrote:
| Violent disagreement is violence.
| mxwsn wrote:
| Good read, thanks for sharing
| esafak wrote:
| Statistical mechanics is the lens that makes most sense to me,
| and it's well studied.
| TechDebtDevin wrote:
| Anyone who wants to demystify ML should read: The StatQuest
| Illustrated Guide to Machine Learning [0] By Josh Starmer.
|
| To this day I haven't found a teacher who could express complex
| ideas as clearly and concisely as Starmer does. It's written in
| an almost children's book like format that is very easy to read
| and understand. He also just published a book on NN that is just
| as good. Highly recommend even if you are already an expert as it
| will give you great ways to teach and communicate complex ideas
| in ML.
|
| [0]: https://www.goodreads.com/book/show/75622146-the-
| statquest-i...
| kaptainscarlet wrote:
| Double Bam
| ajitid wrote:
| Also would like to add that he has a YouTube channel as well
| https://youtube.com/@statquest
| Lerc wrote:
| I have followed a fair few StatQuest and other videos
| (treadmills with Youtube are great for fitness and learning in
| one)
|
| I find that no single source seems to cover things in a way
| that I easily understand, but cumulatively they fill in the
| blanks of each other.
|
| Serrano Academy has been a good source for me as well.
| https://www.youtube.com/@SerranoAcademy/videos
|
| The best tutorials give you a clear sense that the teacher has
| a clear understanding of the underlying principles and how/why
| they are applied.
|
| I have seen a fair few things that are effectively.
|
| _' To do X, you {math thing}'_ While also creating the
| impression that they don't understand why {math thing} is the
| right thing to do, just that {math thing} has a name and it
| produces the result. Meticulously explaining the minutiae of
| {math thing} substitutes for a understanding of what it is
| doing.
|
| It really stood out to me when looking at UMAP and seeing a
| bunch of things where they got into the weeds in the math
| without explaining why these were the particular weeds to be
| looking in.
|
| Then I found a talk by Leland McInnes that had the format.
|
| _{math thing} is a tool to do {objective}. It works, there is
| a proof, you don 't need to understand it to use the tool but
| the info for that is over there if you want to tale a look.
| These are our objectives, let's use these tools to achieve
| them._
|
| The tools are neither magical black boxes, nor confused with
| the actual goal. It really showed the power of fully
| understanding the topic.
| getnormality wrote:
| > rather than restricting the hypothesis space to avoid
| overfitting, embrace a flexible hypothesis space, with a soft
| preference for simpler solutions that are consistent with the
| data. This principle can be encoded in many model classes, and
| thus deep learning is not as mysterious or different from other
| model classes as it might seem.
|
| How does deep learning do this? The last time I was deeply
| involved in machine learning, we used a penalized likelihood
| approach. To find a good model for data, you would optimize a
| cost function over model space, and the cost function was the sum
| of two terms: one quantifying the difference between model
| predictions and data, and the other quantifying the model's
| complexity. This framework encodes exactly a "soft preference for
| simpler solutions that are consistent with the data", but is that
| how deep learning works? I had the impression that the way
| complexity is penalized in deep learning was more complex, less
| straightforward.
| bornfreddy wrote:
| I'm not a guru myself, but I'm sure someone will correct me if
| I'm wrong. :-)
|
| The usual approach to supervised ML is to "invent" the model
| (layers, their parameters) or more often copy one from known
| good reference, then define the cost function and feed it data.
| "Deep" learning just means that instead of a few layers you use
| a big number of them.
|
| What you describe sounds like an automated way of tweaking the
| architecture, IIUC? Never done that, usually the cost of a run
| was too high to let an algorithm do that for me. But I'm
| curious if this approach is being used?
| woopwoop wrote:
| Yeah, it's straightforward to reproduce the results of the
| paper whose conclusion they criticize, "Understanding deep
| learning requires rethinking generalization", without any
| (explicit) regularization or anything else that can be easily
| described as a "soft preference for simpler solutions".
| whiteandnerdy wrote:
| You're correct, and the term you're looking for is
| "regularisation".
|
| There are two common ways of doing this: * L1 or L2
| regularisation: penalises models whose weight matrices are
| complex (in the sense of having lots of large elements) *
| Dropout: train on random subsets of the neurons to force the
| model to rely on simple representations that are distributed
| robustly across its weights
| levocardia wrote:
| Dropout is roughly equivalent to layer-specific L2
| regularization, and it's easy to see why: asymptotically,
| dropping out random neurons will achieve something similar to
| shrinking weights towards zero proportional to their
| (squared) magnitude.
|
| Trevor Hastie's Elements of Statistical Learning has a nice
| proof that (for linear models) L2 regularization is also
| semi-equivalent to dimensionality reduction, which you could
| use to motivate a "simplicity prior" idea in deep learning.
|
| Yet another way of thinking about it, in the context of ReLU
| units, is that a layer of ReLUs forms a truncated hyper-plane
| basis (like splines but in higher dimensions) in feature
| space, and regularization induces smoothness in this
| N-dimensional basis by shrinking that basis towards being a
| flat hyper-plane
| jonathanhuml wrote:
| The solution to the L1 regularization problem is actually a
| specific form of the classical ReLU nonlinearity used in deep
| learning. I'm not sure if similar results hold for other
| nonlinearities, but this gave me good intuition for what
| thresholding is doing mathematically!
| chriskanan wrote:
| Here is an example for data-efficient vision transformers:
| https://arxiv.org/abs/2401.12511
|
| Vision transformers have a more flexible hypothesis space, but
| they tend to have worse sample complexity than convolutional
| networks which have a strong architectural inductive bias. A
| "soft inductive bias" would be something like what this paper
| does where they have a special scheme for initializing vision
| transformers. So schemes like initialization that encourage the
| model to find the right solution without excessively
| constraining it would be a soft preference for simpler
| solutions.
| inciampati wrote:
| An example, which is interesting, in which "deep" networks are
| necessary, is discussed in this fascinating and popular recent
| paper on RNNs [1]. Despite the fact that the minGRU and minLSTM
| models they propose don't explicitly model ordered state
| dependency, they _can_ learn them as long as they are deep enough
| (deep >= 3):
|
| > Instead of explicitly modelling dependencies on previous states
| to capture long-range dependencies, these kinds of recurrent
| models can learn them by stacking multiple layers.
|
| [1] https://arxiv.org/abs/2410.01201
| YesBox wrote:
| I wish I had the time to try this:
|
| 1.) Grab many GBs of text (books, etc).
|
| 2.) For each word, for each next $N words, store distance from
| current word, and increment count for word pair/distance.
|
| 3.) For each word, store most frequent word for each $N distance.
| [a]
|
| 4.) Create a prediction algorithm that determines the next word
| (or set of words) to output from any user input. Basically this
| would compare word pairs/distance and find most probable next set
| of word(s)
|
| How close would this be to GPT 2?
|
| [a] You could go one step further and store multiple words for
| each distance, ordered by frequency
| currymj wrote:
| this is pretty close to how language models worked in the
| 90s-2000s. deep language models -- even GPT 2 -- are much much
| better. on the other hand, the n-gram language models are
| "surprisingly good" even for small n.
| WheatMillington wrote:
| Pretty sure this wouldn't produce anything useful. Pretty sure
| this would generate incoherent gibberish that looks and sounds
| like English but makes no sense. This ignores perhaps the most
| important element of LLM's, the attention mechanism.
| docfort wrote:
| There is some recent work [0] that explores this idea, scaling
| up n-gram models substantially while using word2vec vectors to
| understand similarity. Used to compute something the authors
| call the Creativity Index [1].
|
| [0]: https://infini-gram.io [1]:
| https://arxiv.org/abs/2410.04265v1
| montebicyclelo wrote:
| > How close would this be to GPT 2
|
| Here's a post from 2015 doing something a bit like this [1]
|
| [1] https://nbviewer.org/gist/yoavg/d76121dfde2618422139
| nickysielicki wrote:
| Markov chains are very very far off from gpt2.
| janalsncm wrote:
| The problem is that for any reasonable value of N (>100) you
| will need prohibitive amounts of storage. And it will be
| extremely sparse. And you won't capture any interactions
| between N-99 and N-98.
|
| Transformers do that fairly well and are pretty efficient in
| training.
| 0cf8612b2e1e wrote:
| The scaling is brutal. If you have a 20k word vocabulary and
| want to do 3 grams, you need a 20000^3 matrix of elements (8
| trillion). Most of which is going to be empty.
|
| GPT and friends cheat by not modeling each word separately, but
| a large dimensional "embedding" (just a vector if you also find
| new vocabulary silly). The embedding represents similar words
| near each other in this space. The famous king-man-queen
| example. So even if your training set has never seen "The Queen
| ordered the traitor <blank>", it might have previously seen,
| "The King ordered the traitor beheaded". The vector
| representation lets the model use words that represent similar
| concepts without concrete examples.
| andrewla wrote:
| Importantly, though, LLMs do not take the embeddings as input
| during training; they take the tokens and learn the
| embeddings as part of the training.
|
| Specifically all Transformer-based models; older models used
| things like word2vec or elmo, but all current LLMs train
| their embeddings from scratch.
| naasking wrote:
| And tokens are now going down to the byte level:
|
| https://ai.meta.com/research/publications/byte-latent-
| transf...
| fsndz wrote:
| I actually tried sth like that with the Bible back in 2021.
| scaling is bitch. very difficult to train these types of
| models.
| EncomLab wrote:
| The implication that any software is "mysterious" is problematic
| - there is no "woo" here - the exact state of the machine running
| the software may be determined at every cycle. The exact
| instruction and the data it executed with may be precisely
| determined, as can the next instruction. The entire mythos of any
| software being a "black box" is just so much advertising jargon,
| perpetuated by tech bros who want to believe they are part of
| some Mr. Robot self-styled priestly class.
| margalabargala wrote:
| You're misunderstanding. A level of abstraction is necessary
| for operation of modern systems. There is no human alive who,
| given an intermediate step in the middle of some running
| learning algorithm, is able to understand and mentally model
| the full system at full man-made resolution, that is, down to
| the transistor level, on a modern CPU. Someone wishing to
| understand a piece of software in 2025 is forced to, at some
| point, accept that something somewhere "does what it says on
| the tin" and model it thusly rather than having a full
| understanding.
| EncomLab wrote:
| It's not misunderstanding at all - but your response is
| certainly an attempt to obfuscate the point being made. The
| moment you represent anything in code, you are abstracting a
| real thing into it's digital representation. That digital
| representation if fully formed at every cycle of the digital
| system processing it, and the state of the system - all the
| way down to the transistor level may be precisely determined.
| To say otherwise is to make the same error as those who claim
| that consciousness or understanding are indefinable "extra-
| ordinary" things that we have to just accept exist without
| any justification or evidence.
| margalabargala wrote:
| Okay, then, you're just using your own personal definition
| of "black box" instead of the one everyone else uses.
|
| Something that's a black box is unknown to the speaker.
| It's not understood to be unknowable to anyone.
| 01HNNWZ0MV43FF wrote:
| But the weights trained from machine learning are a black box,
| in the sense that no human designed e.g. the image processing
| kernels that those weights represent.
|
| That is one reason people are skeptical of them, not only is
| training a large model at home expensive, not only is the data
| too big to trivially store, but the weights are not trivial to
| debug either
| xmprt wrote:
| The mystery was never in the "how do computers calculate the
| probabilities of next tokens" but rather in the "why is it able
| to work so well" and "what does this individual neuron
| contribute to the whole model"
| rottc0dd wrote:
| If anyone wants to delve into machine learning, one of the superb
| resources I have found is, Stanfords "Probability for computer sc
| ientists"(https://www.youtube.com/watch?v=2MuDZIAzBMY&list=PLoROM
| vodv4...).
|
| It delves into theoretical underpinnings of probability theory
| and ML, IMO better than any other course I have seen. (Yeah,
| Andrew Ng is legendary, but his course demands some mathematical
| familarity with linear algebra topics)
|
| And of course, for deep learning, 3b1b is great for getting some
| visual introduction (https://www.youtube.com/watch?v=aircAruvnKk&
| list=PLZHQObOWTQ...).
| rottc0dd wrote:
| Caltech's learning from data was really good too, if someone is
| looking for theoretical understanding of ML topics.
|
| https://work.caltech.edu/telecourse
| randomtoast wrote:
| Apparently the word "delve" is the biggest indicator of the use
| of ChatGPT according to Paul Graham.
| wincy wrote:
| I'd love to see an article delve into why that is.
| treyd wrote:
| Because it's common in Nigerian English, which is where
| they outsourced a lot of the RLHF conditioning work to.
| Hugsun wrote:
| Really!? Do you have a source for this? This would be
| really interesting if true.
| sva_ wrote:
| https://pshapira.net/2024/03/31/delving-into-delve/
| EGreg wrote:
| Saying that kind of stuff is the biggest indicator of Paul
| Graham (pg) himself
| tasty_freeze wrote:
| That seems utterly bizarre to me. I don't use "delve"
| frequently myself, but it is common enough that it doesn't
| jump out as an unusual word. Perhaps it is overused or used
| in a not-exactly-usual context that tips one off that it is
| LLM-generated, but by itself it signifies nothing to me.
| godelski wrote:
| It is a very common word used in Nigerian style English
| which was a very common place they were outsourcing RLHF
| tasks to. A sibling comment has a link but it is also easy
| to google.
| chamomeal wrote:
| I watched the 3b1b series on neural nets years ago, and it
| still accounts for 95% of my understanding of AI in general.
|
| I'm not an ML person, but still. That guy has a serious gift
| for explaining stuff.
|
| His video on the uncertainty principle explained stuff to me
| that my entire undergrad education failed to!
| bogeholm wrote:
| Looks nice - are there written versions?
| vcdimension wrote:
| and if anyone is interested in delving more deeply into the
| statistical concepts & results referenced in the paper of this
| post (e.g. VC-dimension, PAC-learning, etc), I can recommend
| this book: https://amzn.eu/d/7Zwe6jw
| buffalobuffalo wrote:
| When I was first getting into Deep Learning, learning the proof
| of the universal approximation theorem helped a lot. Once you
| understand why neural networks are able to approximate functions,
| it makes everything built on top of them much easier to
| understand.
| talles wrote:
| Correct me if I'm wrong, but an artificial neuron is just good
| old linear regression followed by an activation function to make
| it non linear. Make a network out of it and cool stuff happens.
| esafak wrote:
| MLPs are compositions of _generalized_ linear models. That 's
| not very enlightening though; the "mysterious" part is the
| macroscopics of the composition, which you can't really
| understand with the tools of statistics.
| hatthew wrote:
| This is like saying "the human brain is just some chemistry."
| You have the general idea correct, but there's a lot more going
| on that just that, and the emergent system is so much more
| complex that it deserves its own separate field.
| andrewla wrote:
| In a sense; linear regression can be computed exactly so refers
| to a specific technique for producing a linear model.
|
| Most artificial neurons are trained stochastically rather than
| holistically, i.e. rather than looking at the entire training
| set and computing the gradient to minimize the squared loss or
| something similar, they look at each training example and
| compute the local gradient and make small changes in that
| direction.
|
| In addition, the "activation function" almost universally used
| now is the rectified linear unit, which is linear for positive
| input and zero for negative input. This is non-decreasing at
| least as a function, but the fact that it is not monotonic
| means that there is no additional loss accrued for
| overcorrecting in the negative direction.
|
| Given this, using the term "linear regression" to describe the
| model of an artificial neuron is not really a useful heuristic.
| totetsu wrote:
| So where is the line that something becomes 'AI' and is
| regulated?
___________________________________________________________________
(page generated 2025-03-17 23:00 UTC)