[FRIAM] Deep learning training material
Steve Smith
sasmyth at swcp.com
Mon Jan 9 13:31:01 EST 2023
The "celebration of the hand" project (coordinated with UNM's Maxwell
museum of anthropology) is nearly 15 years defunct now. We made a tiny
bit of progress on some LANL small-business-assistance time from an ML
researcher, a small bit of seed-funding and matching time on my own
part. The key to this was the newly available prosumer-grade
laser-scanners of the time, as well as emerging photogrammetric techniques.
The point of the project was to augment what humans were already
doing... both pointing to similarities too subtle for a human to notice
or based on dimensions a human expert wouldn't be able to identify
directly. I haven't touched bases on where that field has gone in most
of that intervening time. My paleontological/archaelogical partner in
that endeavor went his own way (he was more interested in developing a
lifestyle business for himself than actually applying advanced
techniques to the problem at "hand" (pun acknowledged)) and then a few
years later he died which is part of the reason for not (re)visiting it
myself...
The relevant issue to the linked article(s) is mostly that we were
building modelless models (model-free learning) from the data with the
hope (assumption?) that *some* of the features that might emerge as
being good correlates to related/identical "hands" might be ones that
humans could detect or make sense of themselves. The state of the art at
the time was definitely an "art" and was based on the expertise of
contemporary flint knappers trying to reproduce the patterns found in
non-contemporary artifacts. We did not do any actual work (just
speculation guided background research) on pottery and textiles... the
Maxwell museum researcher we worked with was also "overcome by events"
simply needing to develop the exhibits and not having time/focus to
attend to the more researchy aspects... her interest was more on
textiles than ceramics or lithics... which seemed to be the area we
likely had the least advantage over human-experts.
My niece works at the U of Az Archaelogy dept cataloging the
accellerating number of artifacts coming in. At some point I can
imagine an automated classification system taking over much of the
"mundane" aspects of this work... like a self-driving car that at
least provides collision avoidance and lane following...
On 1/9/23 10:29 AM, Pieter Steenekamp wrote:
> Thanks for the references, I've briefly looked at them and am looking
> forward to perusing them more closely. The interpretation of ML is a
> big thing of course. The machine gives you the results and it would
> sometimes be nice if it can be accompanied by some explanation of how
> it achieved it.
>
> What you describe about your work sounds very interesting indeed. My
> gut feel is that, at least with the current generation of AI, human
> ingenuity and judgement would be far superior to AI in recognising
> similarities in the "hand" of the artisans involved. But AI could well
> be a powerful tool in assisting the humans?
>
> On Mon, 9 Jan 2023 at 18:50, Steve Smith <sasmyth at swcp.com> wrote:
>
> I was hoping this article would have more meat to it, but the main
> point seems highly relevant to a practical/introductory workshop
> such as the one you are developing:
>
> The Need for Interpretable Features: Taxonomy and Motivation
> <https://dl.acm.org/doi/10.1145/3544903.3544905>
>
> https://scitechdaily.com/mit-taxonomy-helps-build-explainability-into-the-components-of-machine-learning-models/
>
> My limited experience with ML is that the convo/invo-lutions that
> developers go through to make their learning models work well tend
> to obscure the interpretability of the results. Some of this
> seems unavoidable (inevitable?), particularly the highly technical
> reserved terminology that often exists in a given domain, but this
> team purports to help provide guidelines for minimizing the
> consequences...
>
> My main experience with this domain involved an attempt to find
> distance measures between the flake patterns in 3D scanned
> archaeological artifacts, starting with lithics but also aspiring
> to work with pottery and textiles... essentially trying to
> recognize similarities in the "hand" of the artisans involved.
>
>
> On 1/9/23 1:29 AM, Pieter Steenekamp wrote:
>> @Russ,
>>
>> a) I appreciate the suggestion to include a simple neural network
>> that can make predictions based on inputs and be trained using
>> steepest descent to optimize its weights. This would be a
>> valuable addition to my training material, as it provides a
>> foundational understanding of how neural networks work.
>>
>> b) My focus is on providing practical training for professionals
>> with deep domain knowledge but limited software experience, who
>> are looking to utilize deep learning as a tool in their
>> respective fields. In contrast, it seems that your focus is on
>> understanding the inner workings of deep learning. Both
>> approaches have their own merits, and it is important to cater to
>> the needs and goals of different learners.
>>
>> Pieter
>>
>> On Sun, 8 Jan 2023 at 21:20, Marcus Daniels
>> <marcus at snoutfarm.com> wrote:
>>
>> The main defects of both R and Python are a lack of a typing
>> system and high performance compilation. I find R still
>> follows (is used by) the statistics research community more
>> than Python. Common Lisp was always better than either.
>>
>> Sent from my iPhone
>>
>>> On Jan 8, 2023, at 11:03 AM, Russ Abbott
>>> <russ.abbott at gmail.com> wrote:
>>>
>>>
>>> As indicated in my original reply, my interest in this
>>> project grows from my relative ignorance of Deep Learning.
>>> My career has focussed exclusively on symbolic computing.
>>> I've worked with and taught (a) functional programming,
>>> logic programming, and related issues in advanced Python;
>>> (b) complex systems, agent-based modeling, genetic
>>> algorithms, and related evolutionary processes, (c) a bit of
>>> constraint programming, especially in MiniZinc, and (d)
>>> reinforcement learning as Q-learning, which is reinforcement
>>> learning without neural nets. I've always avoided neural
>>> nets--and more generally numerical programming of any sort.
>>>
>>> Deep learning has produced so many impressive results that
>>> I've decided to devote much of my retirement life to
>>> learning about it. I retired at the end of Spring 2022 and
>>> (after a break) am now devoting much of my time to learning
>>> more about Deep Neural Nets. So far, I've dipped my brain
>>> into it at various points. I think I've learned a fair
>>> amount. For example,
>>>
>>> * I now know how to build a neural net (NN) that adds two
>>> numbers using a single layer with a single neuron. It's
>>> really quite simple and is, I think, a beautiful example
>>> of how NNs work. If I were to teach an intro to NNs I'd
>>> start with this.
>>> * I've gone through the Kaggle Deep Learning sequence
>>> mentioned earlier.
>>> * I found a paper that shows how you can approximate
>>> any differentiable function to any degree of accuracy
>>> with a single-layer NN. (This is a very nice result,
>>> although I believe it's not used explicitly in building
>>> serious Deep NN systems.)
>>> * From what I've seen so far, most serious DNNs are built
>>> using Keras rather than PyTorch.
>>> * I've looked at Jeremy Howard's fast.ai <http://fast.ai>
>>> material. I was going to go through the course but
>>> stopped when I found that it uses PyTorch. Also, it
>>> seems to be built on fast.ai <http://fast.ai> libraries
>>> that do a lot of the work for you without explanation.
>>> And it seems to focus almost exclusively on
>>> Convolutional NNs.
>>> * My impression of DNNs is that to a great extent they are
>>> /ad hoc/. There is no good way to determine the best
>>> architecture to use for a given problem. By
>>> architecture, I mean the number of layers, the number of
>>> neurons in each layer, the types of layers, the
>>> activation functions to use, etc.
>>> * All DNNs that I've seen use Python as code glue rather
>>> than R or some other language. I like Python--so I'm
>>> pleased with that.
>>> * To build serious NNs one should learn the Python
>>> libraries Numpy (array manipulation) and Pandas (data
>>> processing). Numpy especially seems to be used for
>>> virtually all DNNs that I've seen.
>>> * Keras and probably PyTorch include a number of
>>> special-purpose neurons and layers that can be
>>> included in one's DNN. These include: a DropOut layer,
>>> LSTM (short-long-term memory) neurons, convolutional
>>> layers, recurrent neural net layers (RNN), and more
>>> recently transformers, which get credit for ChatGPT and
>>> related programs. My impression is that these
>>> special-purpose layers are /ad hoc/ in the same sense
>>> that functions or libraries that one finds useful in a
>>> programming language are /ad hoc/. They have been very
>>> important for the success of DNNs, but they came into
>>> existence because people invented them in the same way
>>> that people invented useful functions and libraries.
>>> * NN libraries also include a menagerie of activation
>>> functions. An activation function acts as the final
>>> control on the output of a layer. Different
>>> activation functions are used for different purposes. To
>>> be successful in building a DNN, one must understand
>>> what those activation functions do for you and which
>>> ones to use.
>>> * I'm especially interested in DNNs that use reinforcement
>>> learning. That's because the first DNN work that
>>> impressed me was DeepMind's DNNs that learned to play
>>> Atari games--and then Go, etc. An important advantage of
>>> Reinforcement Learning (RL) is that it doesn't depend on
>>> mountains of labeled data.
>>> * I find RL systems more interesting than image
>>> recognition systems. One of the striking features of
>>> many image recognition systems is that they can be
>>> thrown off by changing a small number of pixels in an
>>> image. The changed image would look to a human observer
>>> just like the original, but it might fool a trained NN
>>> into labeling the image as a banana rather than, say, an
>>> automobile, which is what it really is. To address this
>>> problem people have developed Generative Adversarial
>>> Networks (GANs) which attempt to find such weaknesses in
>>> a neural net during training and then to train the NN
>>> not to have those weaknesses. This is a fascinating
>>> result, but as far as I can tell, it mainly shows how
>>> fragile some NNs are and doesn't add much conceptual
>>> depth to one's understanding of how NNs work.
>>>
>>> I'm impressed with this list of things I sort of know. If
>>> you had asked me before I started writing this email I
>>> wouldn't have thought I had learned as much as I have. Even
>>> so, I feel like I don't understand much of it beyond a
>>> superficial level.
>>>
>>> So far I've done all my exploration using Google's Colab
>>> (Google's Python notebook implementation) and Kaggle's
>>> similar Python notebook implementation. (I prefer Colab to
>>> Kaggle.) Using either one, it's super nice not to have to
>>> download and install anything!
>>>
>>> I'm continuing my journey to learn more about DNNs. I'd be
>>> happy to have company and to help develop materials to teach
>>> about DNNs. (Developing teaching materials always helps me
>>> learn the subject being covered.)
>>>
>>> -- Russ Abbott
>>> Professor Emeritus, Computer Science
>>> California State University, Los Angeles
>>>
>>>
>>> On Sun, Jan 8, 2023 at 1:48 AM glen <gepropella at gmail.com>
>>> wrote:
>>>
>>> Yes, the money/expertise bar is still pretty high. But
>>> TANSTAAFL still applies. And the overwhelming evidence
>>> is coming in that specific models do better than those
>>> trained up on diverse data sets, "better" meaning less
>>> prone to subtle bullsh¡t. What I find fascinating is
>>> tools like OpenAI *facilitate* trespassing. We have a
>>> wonderful bloom of non-experts claiming they understand
>>> things like "deep learning". But do they? An old
>>> internet meme is brought to mind: "Do you even Linear
>>> Algebra, bro?" >8^D
>>>
>>> On 1/8/23 01:06, Jochen Fromm wrote:
>>> > I have finished a number of Coursera courses recently,
>>> including "Deep Learning & Neural Networks with Keras"
>>> which was ok but not great. The problems with deep
>>> learning are
>>> >
>>> > * to achieve impressive results like chatGPT from
>>> OpenAi or LaMDA from Goggle you need to spend millions
>>> on hardware
>>> > * only big organisations can afford to create such
>>> expensive models
>>> > * the resulting network is s black box and it is
>>> unclear why it works the way it does
>>> >
>>> > In the end it is just the same old back propagation
>>> that has been known for decades, just on more computers
>>> and trained on more data. Peter Norvig calls it "The
>>> unreasonable effectiveness of data"
>>> > https://research.google.com/pubs/archive/35179.pdf
>>> >
>>> > -J.
>>> >
>>> >
>>> > -------- Original message --------
>>> > From: Russ Abbott <russ.abbott at gmail.com>
>>> > Date: 1/8/23 12:20 AM (GMT+01:00)
>>> > To: The Friday Morning Applied Complexity Coffee Group
>>> <friam at redfish.com>
>>> > Subject: Re: [FRIAM] Deep learning training material
>>> >
>>> > Hi Pieter,
>>> >
>>> > A few comments.
>>> >
>>> > * Much of the actual deep learning material looks
>>> like it came from the Kaggle "Deep Learning
>>> <https://www.kaggle.com/learn/intro-to-deep-learning>"
>>> sequence.
>>> > * In my opinion, R is an ugly and /ad hoc/ language.
>>> I'd stick to Python.
>>> > * More importantly, I would put the
>>> How-to-use-Python stuff into a preliminary class. Assume
>>> your audience knows how to use Python and focus on Deep
>>> Learning. Given that, there is only a minimal amount of
>>> information about Deep Learning in the write-up. If I
>>> were to attend the workshop and thought I would be
>>> learning about Deep Learning, I would be
>>> disappointed--at least with what's covered in the write-up.
>>> >
>>> > I say this because I've been looking for a good
>>> intro to Deep Learning. Even though I taught Computer
>>> Science for many years, and am now retired, I avoided
>>> Deep Learning because it was so non-symbolic. My focus
>>> has always been on symbolic computing. But Deep Learning
>>> has produced so many extraordinarily impressive results,
>>> I decided I should learn more about it. I haven't found
>>> any really good material. If you are interested, I'd be
>>> more than happy to work with you on developing some
>>> introductory Deep Learning material.
>>> >
>>> > -- Russ Abbott
>>> > Professor Emeritus, Computer Science
>>> > California State University, Los Angeles
>>> >
>>> >
>>> > On Thu, Jan 5, 2023 at 11:31 AM Pieter Steenekamp
>>> <pieters at randcontrols.co.za
>>> <mailto:pieters at randcontrols.co.za>> wrote:
>>> >
>>> > Thanks to the kind support of OpenAI's chatGPT, I
>>> am in the process of gathering materials for a
>>> comprehensive and hands-on deep learning workshop.
>>> Although it is still a work in progress, I welcome any
>>> interested parties to take a look and provide their
>>> valuable input. Thank you!
>>> >
>>> > You can get it from:
>>> >
>>> https://www.dropbox.com/s/eyx4iumb0439wlx/deep%20learning%20training%20rev%2005012023.zip?dl=0
>>> <https://www.dropbox.com/s/eyx4iumb0439wlx/deep%20learning%20training%20rev%2005012023.zip?dl=0>
>>> >
>>>
>>> --
>>> ꙮ Mɥǝu ǝlǝdɥɐuʇs ɟᴉƃɥʇ' ʇɥǝ ƃɹɐss snɟɟǝɹs˙ ꙮ
>>>
>>> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-.
>>> --- -.. .
>>> FRIAM Applied Complexity Group listserv
>>> Fridays 9a-12p Friday St. Johns Cafe / Thursdays
>>> 9a-12p Zoom https://bit.ly/virtualfriam
>>> to (un)subscribe
>>> http://redfish.com/mailman/listinfo/friam_redfish.com
>>> FRIAM-COMIC http://friam-comic.blogspot.com/
>>> archives: 5/2017 thru present
>>> https://redfish.com/pipermail/friam_redfish.com/
>>> 1/2003 thru 6/2021 http://friam.383.s1.nabble.com/
>>>
>>> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. ---
>>> -.. .
>>> FRIAM Applied Complexity Group listserv
>>> Fridays 9a-12p Friday St. Johns Cafe / Thursdays 9a-12p
>>> Zoom https://bit.ly/virtualfriam
>>> to (un)subscribe
>>> http://redfish.com/mailman/listinfo/friam_redfish.com
>>> FRIAM-COMIC http://friam-comic.blogspot.com/
>>> archives: 5/2017 thru present
>>> https://redfish.com/pipermail/friam_redfish.com/
>>> 1/2003 thru 6/2021 http://friam.383.s1.nabble.com/
>> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. ---
>> -.. .
>> FRIAM Applied Complexity Group listserv
>> Fridays 9a-12p Friday St. Johns Cafe / Thursdays 9a-12p
>> Zoom https://bit.ly/virtualfriam
>> to (un)subscribe
>> http://redfish.com/mailman/listinfo/friam_redfish.com
>> FRIAM-COMIC http://friam-comic.blogspot.com/
>> archives: 5/2017 thru present
>> https://redfish.com/pipermail/friam_redfish.com/
>> 1/2003 thru 6/2021 http://friam.383.s1.nabble.com/
>>
>>
>> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
>> FRIAM Applied Complexity Group listserv
>> Fridays 9a-12p Friday St. Johns Cafe / Thursdays 9a-12p Zoomhttps://bit.ly/virtualfriam
>> to (un)subscribehttp://redfish.com/mailman/listinfo/friam_redfish.com
>> FRIAM-COMIChttp://friam-comic.blogspot.com/
>> archives: 5/2017 thru presenthttps://redfish.com/pipermail/friam_redfish.com/
>> 1/2003 thru 6/2021http://friam.383.s1.nabble.com/
> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
> FRIAM Applied Complexity Group listserv
> Fridays 9a-12p Friday St. Johns Cafe / Thursdays 9a-12p Zoom
> https://bit.ly/virtualfriam
> to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
> FRIAM-COMIC http://friam-comic.blogspot.com/
> archives: 5/2017 thru present
> https://redfish.com/pipermail/friam_redfish.com/
> 1/2003 thru 6/2021 http://friam.383.s1.nabble.com/
>
>
> -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
> FRIAM Applied Complexity Group listserv
> Fridays 9a-12p Friday St. Johns Cafe / Thursdays 9a-12p Zoomhttps://bit.ly/virtualfriam
> to (un)subscribehttp://redfish.com/mailman/listinfo/friam_redfish.com
> FRIAM-COMIChttp://friam-comic.blogspot.com/
> archives: 5/2017 thru presenthttps://redfish.com/pipermail/friam_redfish.com/
> 1/2003 thru 6/2021http://friam.383.s1.nabble.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://redfish.com/pipermail/friam_redfish.com/attachments/20230109/100f6ee4/attachment-0001.html>
More information about the Friam
mailing list