youwen5/alexandria
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

udpate

Browse source
This commit is contained in:
Youwen Wu 2024-11-20 14:51:30 -08:00
parent c8229c56b8
commit ac0354cb42
Signed by: youwen5
GPG key ID: 865658ED1FE61EC3
7 changed files with 815 additions and 55 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

164
2024/documents/by-course/phil-1/paper-2/main.typ
View file

@ -32,10 +32,6 @@
],
[
Word Count: #total-words
#footnote[
Figure computed programmatically during document compilation. Discounts
content in tables and the AI contribution statement.
]
],
),
),
@ -43,12 +39,12 @@
= Introduction
The argument Against Fearing Death says that you should not fear being dead
In _Against Fearing Death_, the author argues you should not fear being dead
because it is not bad for you. In this paper, I reject this thesis by showing
that the argument from hedonism with which the author supports their premises
forces us to draw incorrect conclusions.
that the argument from hedonism the author relies on forces us to draw the
absurd conclusion that being alive is worse than being dead.
The author states the argument against fearing death as follows on
The author states the argument Against Fearing Death as follows on
#cite(<Korman2022-KORLFA>, supplement: [p. 74]):
#indented-argument(
@ -70,31 +66,35 @@ under anesthesia during a risky surgery, you should not fear that either, as
you will painlessly transition from being alive to being dead, and there is
nothing to fear about being dead.
The author justifies FD1 by denying the only possibility of rejecting it, which
is the notion that your consciousness continues after the death of your
physical body, perhaps in an afterlife. He rejects this idea by arguing that
you, the consciousness reading this, and your physical human body, are one and
the same (i.e. they are _numerically equivalent_). So, if your physical human
body biologically ceases to be conscious after death (which it does), then you
also cease to be conscious when you die.
The author justifies FD1 on #cite(<Korman2022-KORLFA>, supplement: [pp. 79-82])
by denying the only possibility of rejecting it, which is the notion that your
consciousness continues after the death of your physical body (perhaps in an
afterlife). He rejects this idea by arguing that you, the consciousness reading
this, and your physical human body, are one and the same (i.e. they are
_numerically equivalent_). So, if your physical human body ceases to be
conscious after death (which it does), then you (the conscious being reading
this) also cease to be conscious when you die.
Premise FD3 is the direct result of FD1 and FD2, so its validity is dependent
entirely on FD1 and FD2. We've already shown why FD1 should be accepted. Let us
first examine FD4 before returning to FD2.
Premise FD3 is the direct conclusion of FD1 and FD2, so it depends on the
validity of those premises. We've already shown why FD1 should be accepted. Let
us first examine FD4 before taking a closer look at FD2.
FD4 makes a lot of intuitive sense. Indeed, upon close inspection, there are no
glaring issues. It is rational to fear something if and only if it is bad for
you. Sure, you may fear things that are not bad for you, but these fears are
_irrational_, you should try to resist them. The only things that it is
rational to fear are things that are actually bad for you, such as burning your
hand on a hot stove, or stubbing your toe.
FD4 makes a lot of intuitive sense, and upon closer inspection, there are
indeed no issues which arise. It is irrational to fear something if it is not
bad for you (this does not imply everything bad for you should necessarily be
feared). Sure, you _could_ fear things that are not bad for you, but as the
author states on #cite(<Korman2022-KORLFA>, supplement: [pp. 82-83]), these
fears are _irrational_, and you should try to resist them. The only things you
should fear are things that are actually bad for you, such as burning your hand
on a hot stove, or stubbing your toe.
Finally, we return to premise FD2. This is perhaps the most questionable
premise as it is not immediately clear why being unconscious implies that being
dead is not bad.
Finally, we return to premise FD2. This is arguably the most dubious premise as
it is not immediately clear why being unconscious implies that being dead is
not bad.
We first need a rigorous account of what exactly characterizes something as
"bad." The author defines the following hedonist principle:
"bad." The author defines the following _hedonist principle_ on
#cite(<Korman2022-KORLFA>, supplement: [p. 76]):
#pad(
left: 16pt,
@ -118,10 +118,9 @@ FD2.
$<==>$ (FD2) So, if you cease to be conscious when you die, then being dead isn't bad for you
],
)
I've slightly modified the author's numberings to emphasize that the conclusion
AH3 is equivalent to the premise FD2. That is, if the #smallcaps[Argument from
Hedonism] holds, then FD2 must be true.
Hedonism] holds, then FD2 must also be true.
We've shown that the author's argument for why you should not fear death is
substantiated by what appear to be valid premises. I object in section 2 by
@ -132,15 +131,17 @@ responses to my objection.
= Don't kill yourself
I will advance the argument that we should reject HD\* because its logical
conclusion is that you should kill yourself. HD\* implies that being dead is
not bad for you, but it also implies that _being alive_ is bad for you.
I advance the argument that we should reject HD\* because its logical
conclusion is not only that being dead is not bad for you, but _being alive_ is
actually bad for you. Accordingly, this implies that you should kill yourself
to stop being alive because it is bad for you. Of course, this is wrong and we
should not accept HD\*. Consider the following argument:
#indented-argument(
title: "Argument for Killing Yourself",
abbreviation: "KYS",
[You are occasionally conscious when you are alive],
[If you are conscious, you experience more pain than you otherwise would have if you were unconscious],
[If you are occasionally conscious, you will experience more pain than you otherwise would have if you were unconscious],
[$<==>$ (HD\*) Something is bad for you if and only if it results in more pain than you otherwise would have had],
[So, being alive is bad for you],
[If you are unconscious when you are dead, then being dead isn't bad for you],
@ -148,37 +149,90 @@ not bad for you, but it also implies that _being alive_ is bad for you.
[So, you should stop being alive],
)
In other words, KYS7 plainly states that you should kill yourself to stop being
alive. This is clearly a ridiculous conclusion and one we should not accept. In
order to reject this conclusion, we must reject KYS3, which is HD\*. All of the
other premises can be substantiated easily, as follows.
In other words, KYS7 states that you should find a way to kill yourself in
order to stop being alive. This is an absurd conclusion we should not accept,
and it indicates a serious error with one of our premise. Let us identify
exactly which one went wrong.
KYS1 is trivial. KYS2 is true because you experience no pain when unconscious,
and we certainly experience pain at some point while conscious. So, we must
experience more pain while conscious than we otherwise would have (while
unconscious).
KYS1 is trivial (unless you are unconscious for the rest of your life, which
for our purposes is essentially the same as death).
KYS2 is true because you experience no pain when unconscious, and we certainly
experience pain at some point while conscious. So, we must experience more pain
while conscious than we otherwise would have (while unconscious).
KYS5 is equivalent to our conclusion AH3 in the #smallcaps[Argument from
Hedonism].
KYS6 essentially just says that if being alive is bad and being dead is good,
then you should take action to stop being alive and start being dead. After
all, why continue doing something that is bad for you when the alternative not
bad for you?
all, why continue doing something that is bad for you when the alternative is
not bad for you?
Clearly, our own avenue forward is to reject HD\* as our principle of hedonism.
Without HD\*, the #smallcaps[Argument from Hedonism] no longer stands, and
therefore the argument for FD2 fails.
Clearly, our only option is to reject KYS3 (which is just HD\*) as our
principle of hedonism. Without HD\*, the #smallcaps[Argument from Hedonism] no
longer stands, and therefore the argument for FD2 fails.
Note that we do not make a claim as to whether or not death is bad for you. We
simply show the absurdity of an argument that relies on HD\*, which means the
#smallcaps[Argument from Hedonism] fails to justify the premise FD2. Without a
clear justification for why we should accept the dubious claim in FD2, we can
no longer claim that death is definitely not bad and should not be feared.
// #let hdp = [HD$'$]
//
// Let us formulate a new hedonist principle, denoted #hdp.
//
// #pad(
// left: 16pt,
// [
// (#hdp) Something is bad for you if and only if it prevents or hinders the achievement of your goals
// ],
// )
//
// We need to clarify what is meant by "goals". In this case, goals refers broadly
// to all of the things someone needs to feel fulfilled. Someone may have a few
// _fundamental goals_, such as to be fulfilled or to feel happy.
//
// Anything that works against these goals is bad. If someone seeks to be happy,
// then feeling pain is bad for them because they no longer .
//
// The author's hypothetical case of #smallcaps[Unread Mail] shows that this
// HD\*\* wrongfully characterizes some situations as bad, so we should prefer
// HD\*.
//
// However, in my formulation #hdp, we do correctly identify that the situation in
// #smallcaps[Unread Mail] is not bad. #hdp is essentially equivalent to HD\* is
// most cases, but it successfully identifies that being alive is not bad for you.
// Since being
= Possible objections
One might attempt to object to KYS without rejecting HD\*. The only other
premises to object against are KYS1 and KYS2. In particular, one might raise
the concern that someone could be alive without ever being conscious (as an
objection to KYS1), and that someone may not experience any pain while
conscious (an objection to KYS2).
premises to reasonably object against are KYS1 and KYS2. In particular, one
might raise the following concerns
These criticisms are not substantial enough for us to reject KYS. KYS1 brings
up a valid point, but being unconscious for the rest of your life is not really
substantially different from being dead. Regardless, we could modify our
argument to
+ An objection to KYS1: someone could be alive without ever being conscious.
+ An objection to KYS2: someone may not ever experience any pain while conscious.
These criticisms are not really substantial enough for us to reject KYS. KYS1
brings up a valid point, but being unconscious for the rest of your life is not
really a better situation than being dead. Regardless, HD\* still implies that
being conscious is bad while being in a _death-like state_ of permanent
unconsciousness is not. The conclusion then becomes that you should either kill
yourself or place yourself into a death-like state (perhaps a coma), which is
just as absurd as before.
One may argue from a hypothetical situation in which a human is somehow
modified to be incapable of feeling pain (of any sort). In this situation, HD\*
does not fail, as it does not imply being alive and conscious is worse than
being dead, since being conscious and being dead both result in absolutely no
pain.
This case fails to present any challenge to our argument. Even though HD\* does
not fail in the hypothetical, it clearly still fails _now_, as it still implies
that being alive is bad for you, the person reading, who almost certainly does
feel pain.
#pagebreak()

5
2024/documents/by-name/vision-whitepaper/appendix.typ Normal file
View file

@ -0,0 +1,5 @@
= Appendix / supplemental material
Optionally include supplemental material (complete proofs, additional
experiments and plots) in appendix. All such materials *SHOULD be included in
the main submission*.

524
2024/documents/by-name/vision-whitepaper/checklist.typ Normal file
View file

@ -0,0 +1,524 @@
#import "@preview/bloated-neurips:0.5.1": url
#let TODO = text(fill: red, [*[TODO]*])
#let answerNA = text(fill: gray, "[NA]")
#let answerNo = text(fill: red, "[NO]")
#let answerYes = text(fill: blue, "[YES]")
#pagebreak(weak: true)
#set heading(numbering: none)
= NeurIPS Paper Checklist
#let claim(
name: [],
question: [],
answer: [],
justification: [],
guidelines: [],
) = {
set list(indent: 1em, tight: false)
show list: set block(spacing: 10pt)
set par(spacing: 5.8pt)
[
*#name*
Question: #question
Answer: #answer
Justification: #justification
Guidelines:
#guidelines
]
}
+ #claim(
name: [Claims],
question: [
Do the main claims made in the abstract and introduction accurately reflect
the paper's contributions and scope?],
answer: TODO, // Replace by answerYes, answerNo, or answerNA.
justification: TODO,
guidelines: [
- The answer NA means that the abstract and introduction do not include the
claims made in the paper.
- The abstract and/or introduction should clearly state the claims made,
including the contributions made in the paper and important assumptions and
limitations. A No or NA answer to this question will not be perceived well
by the reviewers.
- The claims made should match theoretical and experimental results, and
reflect how much the results can be expected to generalize to other settings.
- It is fine to include aspirational goals as motivation as long as it is
clear that these goals are not attained by the paper.
])
+ #claim(
name: [Limitations],
question: [
Does the paper discuss the limitations of the work performed by the
authors?
],
answer: TODO, // Replace by answerYes, answerNo, or answerNA.
justification: TODO,
guidelines: [
- The answer NA means that the paper has no limitation while the answer No
means that the paper has limitations, but those are not discussed in the
paper.
- The authors are encouraged to create a separate "Limitations" section in
their paper.
- The paper should point out any strong assumptions and how robust the
results are to violations of these assumptions (e.g., independence
assumptions, noiseless settings, model well-specification, asymptotic
approximations only holding locally). The authors should reflect on how
these assumptions might be violated in practice and what the implications
would be.
- The authors should reflect on the scope of the claims made, e.g., if the
approach was only tested on a few datasets or with a few runs. In
general, empirical results often depend on implicit assumptions, which
should be articulated.
- The authors should reflect on the factors that influence the performance
of the approach. For example, a facial recognition algorithm may perform
poorly when image resolution is low or images are taken in low lighting.
Or a speech-to-text system might not be used reliably to provide closed
captions for online lectures because it fails to handle technical jargon.
- The authors should discuss the computational efficiency of the proposed
algorithms and how they scale with dataset size.
- If applicable, the authors should discuss possible limitations of their
approach to address problems of privacy and fairness.
- While the authors might fear that complete honesty about limitations
might be used by reviewers as grounds for rejection, a worse outcome
might be that reviewers discover limitations that aren't acknowledged in
the paper. The authors should use their best judgment and recognize that
individual actions in favor of transparency play an important role in
developing norms that preserve the integrity of the community. Reviewers
will be specifically instructed to not penalize honesty concerning
limitations.
])
+ #claim(
name: [Theory Assumptions and Proofs],
question: [
For each theoretical result, does the paper provide the full set of
assumptions and a complete (and correct) proof?
],
answer: TODO, // Replace by answerYes, answerNo, or answerNA.
justification: TODO,
guidelines: [
- The answer NA means that the paper does not include theoretical results.
- All the theorems, formulas, and proofs in the paper should be numbered
and cross-referenced.
- All assumptions should be clearly stated or referenced in the statement
of any theorems.
- The proofs can either appear in the main paper or the supplemental
material, but if they appear in the supplemental material, the authors
are encouraged to provide a short proof sketch to provide intuition.
- Inversely, any informal proof provided in the core of the paper should be
complemented by formal proofs provided in appendix or supplemental
material.
- Theorems and Lemmas that the proof relies upon should be properly
referenced.
])
+ #claim(
name: [Experimental Result Reproducibility],
question: [
Does the paper fully disclose all the information needed to reproduce the
main experimental results of the paper to the extent that it affects the
main claims and/or conclusions of the paper (regardless of whether the code
and data are provided or not)?
],
answer: TODO, // Replace by answerYes, answerNo, or answerNA.
justification: TODO,
guidelines: [
- The answer NA means that the paper does not include experiments.
- If the paper includes experiments, a No answer to this question will not
be perceived well by the reviewers: Making the paper reproducible is
important, regardless of whether the code and data are provided or not.
- If the contribution is a dataset and/or model, the authors should
describe the steps taken to make their results reproducible or verifiable.
- Depending on the contribution, reproducibility can be accomplished in
various ways. For example, if the contribution is a novel architecture,
describing the architecture fully might suffice, or if the contribution
is a specific model and empirical evaluation, it may be necessary to
either make it possible for others to replicate the model with the same
dataset, or provide access to the model. In general. releasing code and
data is often one good way to accomplish this, but reproducibility can
also be provided via detailed instructions for how to replicate the
results, access to a hosted model (e.g., in the case of a large language
model), releasing of a model checkpoint, or other means that are
appropriate to the research performed.
- While NeurIPS does not require releasing code, the conference does
require all submissions to provide some reasonable avenue for
reproducibility, which may depend on the nature of the contribution. For
example
#set enum(numbering: "(a)")
+ If the contribution is primarily a new algorithm, the paper should make
it clear how to reproduce that algorithm.
+ If the contribution is primarily a new model architecture, the paper
should describe the architecture clearly and fully.
+ If the contribution is a new model (e.g., a large language model), then
there should either be a way to access this model for reproducing the
results or a way to reproduce the model (e.g., with an open-source
dataset or instructions for how to construct the dataset).
+ We recognize that reproducibility may be tricky in some cases, in which
case authors are welcome to describe the particular way they provide
for reproducibility. In the case of closed-source models, it may be
that access to the model is limited in some way (e.g., to registered
users), but it should be possible for other researchers to have some
path to reproducing or verifying the results.
])
+ #claim(
name: [Open access to data and code],
question: [
Does the paper provide open access to the data and code, with sufficient
instructions to faithfully reproduce the main experimental results, as
described in supplemental material?
],
answer: TODO, // Replace by answerYes, answerNo, or answerNA.
justification: TODO,
guidelines: [
- The answer NA means that paper does not include experiments requiring
code.
- Please see the NeurIPS code and data submission guidelines
(#url("https://nips.cc/public/guides/CodeSubmissionPolicy")) for more
details.
- While we encourage the release of code and data, we understand that this
might not be possible, so "No" is an acceptable answer. Papers cannot be
rejected simply for not including code, unless this is central to the
contribution (e.g., for a new open-source benchmark).
- The instructions should contain the exact command and environment needed
to run to reproduce the results. See the NeurIPS code and data submission
guidelines (#url("https://nips.cc/public/guides/CodeSubmissionPolicy"))
for more details.
- The authors should provide instructions on data access and preparation,
including how to access the raw data, preprocessed data, intermediate
data, and generated data, etc.
- The authors should provide scripts to reproduce all experimental results
for the new proposed method and baselines. If only a subset of
experiments are reproducible, they should state which ones are omitted
from the script and why.
- At submission time, to preserve anonymity, the authors should release
anonymized versions (if applicable).
- Providing as much information as possible in supplemental material
(appended to the paper) is recommended, but including URLs to data and
code is permitted.
])
+ #claim(
name: [Experimental Setting/Details],
question: [
Does the paper specify all the training and test details (e.g., data
splits, hyperparameters, how they were chosen, type of optimizer, etc.)
necessary to understand the results?
],
answer: TODO, // Replace by answerYes, answerNo, or answerNA.
justification: TODO,
guidelines: [
- The answer NA means that the paper does not include experiments.
- The experimental setting should be presented in the core of the paper to
a level of detail that is necessary to appreciate the results and make
sense of them.
- The full details can be provided either with the code, in appendix, or as
supplemental material.
])
+ #claim(
name: [Experiment Statistical Significance],
question: [
Does the paper report error bars suitably and correctly defined or other
appropriate information about the statistical significance of the
experiments?
],
answer: TODO, // Replace by answerYes, answerNo, or answerNA.
justification: TODO,
guidelines: [
- The answer NA means that the paper does not include experiments.
- The authors should answer "Yes" if the results are accompanied by error
bars, confidence intervals, or statistical significance tests, at least
for the experiments that support the main claims of the paper.
- The factors of variability that the error bars are capturing should be
clearly stated (for example, train/test split, initialization, random
drawing of some parameter, or overall run with given experimental
conditions).
- The method for calculating the error bars should be explained (closed
form formula, call to a library function, bootstrap, etc.)
- The assumptions made should be given (e.g., Normally distributed errors).
- It should be clear whether the error bar is the standard deviation or the
standard error of the mean.
- It is OK to report 1-sigma error bars, but one should state it. The
authors should preferably report a 2-sigma error bar than state that they
have a 96% CI, if the hypothesis of Normality of errors is not verified.
- For asymmetric distributions, the authors should be careful not to show
in tables or figures symmetric error bars that would yield results that
are out of range (e.g. negative error rates).
- If error bars are reported in tables or plots, The authors should explain
in the text how they were calculated and reference the corresponding
figures or tables in the text.
])
+ #claim(
name: [Experiments Compute Resources],
question: [
For each experiment, does the paper provide sufficient information on the
computer resources (type of compute workers, memory, time of execution)
needed to reproduce the experiments?
],
answer: TODO, // Replace by answerYes, answerNo, or answerNA.
justification: TODO,
guidelines: [
- The answer NA means that the paper does not include experiments.
- The paper should indicate the type of compute workers CPU or GPU,
internal cluster, or cloud provider, including relevant memory and storage.
- The paper should provide the amount of compute required for each of the
individual experimental runs as well as estimate the total compute.
- The paper should disclose whether the full research project required more
compute than the experiments reported in the paper (e.g., preliminary or
failed experiments that didn't make it into the paper).
])
+ #claim(
name: [Code Of Ethics],
question: [
Does the research conducted in the paper conform, in every respect, with
the NeurIPS Code of Ethics
#url("https://neurips.cc/public/EthicsGuidelines")
],
answer: TODO, // Replace by answerYes, answerNo, or answerNA.
justification: TODO,
guidelines: [
- The answer NA means that the authors have not reviewed the NeurIPS Code
of Ethics.
- If the authors answer No, they should explain the special circumstances
that require a deviation from the Code of Ethics.
- The authors should make sure to preserve anonymity (e.g., if there is a
special consideration due to laws or regulations in their jurisdiction).
])
+ #claim(
name: [Broader Impacts],
question: [
Does the paper discuss both potential positive societal impacts and
negative societal impacts of the work performed?
],
answer: TODO, // Replace by answerYes, answerNo, or answerNA.
justification: TODO,
guidelines: [
- The answer NA means that there is no societal impact of the work
performed.
- If the authors answer NA or No, they should explain why their work has no
societal impact or why the paper does not address societal impact.
- Examples of negative societal impacts include potential malicious or
unintended uses (e.g., disinformation, generating fake profiles,
surveillance), fairness considerations (e.g., deployment of technologies
that could make decisions that unfairly impact specific groups), privacy
considerations, and security considerations.
- The conference expects that many papers will be foundational research and
not tied to particular applications, let alone deployments. However, if
there is a direct path to any negative applications, the authors should
point it out. For example, it is legitimate to point out that an
improvement in the quality of generative models could be used to generate
deepfakes for disinformation. On the other hand, it is not needed to
point out that a generic algorithm for optimizing neural networks could
enable people to train models that generate Deepfakes faster.
- The authors should consider possible harms that could arise when the
technology is being used as intended and functioning correctly, harms
that could arise when the technology is being used as intended but gives
incorrect results, and harms following from (intentional or
unintentional) misuse of the technology.
- If there are negative societal impacts, the authors could also discuss
possible mitigation strategies (e.g., gated release of models, providing
defenses in addition to attacks, mechanisms for monitoring misuse,
mechanisms to monitor how a system learns from feedback over time,
improving the efficiency and accessibility of ML).
])
+ #claim(
name: [Safeguards],
question: [
Does the paper describe safeguards that have been put in place for
responsible release of data or models that have a high risk for misuse
(e.g., pretrained language models, image generators, or scraped datasets)?
],
answer: TODO, // Replace by answerYes, answerNo, or answerNA.
justification: TODO,
guidelines: [
- The answer NA means that the paper poses no such risks.
- Released models that have a high risk for misuse or dual-use should be
released with necessary safeguards to allow for controlled use of the
model, for example by requiring that users adhere to usage guidelines or
restrictions to access the model or implementing safety filters.
- Datasets that have been scraped from the Internet could pose safety
risks. The authors should describe how they avoided releasing unsafe
images.
- We recognize that providing effective safeguards is challenging, and many
papers do not require this, but we encourage authors to take this into
account and make a best faith effort.
])
+ #claim(
name: [Licenses for existing assets],
question: [
Are the creators or original owners of assets (e.g., code, data, models),
used in the paper, properly credited and are the license and terms of use
explicitly mentioned and properly respected?
],
answer: TODO, // Replace by answerYes, answerNo, or answerNA.
justification: TODO,
guidelines: [
- The answer NA means that the paper does not use existing assets.
- The authors should cite the original paper that produced the code package
or dataset.
- The authors should state which version of the asset is used and, if
possible, include a URL.
- The name of the license (e.g., CC-BY 4.0) should be included for each
asset.
- For scraped data from a particular source (e.g., website), the copyright
and terms of service of that source should be provided.
- If assets are released, the license, copyright information, and terms of
use in the package should be provided. For popular datasets,
#url("https://paperswithcode.com/datasets") has curated licenses for some
datasets. Their licensing guide can help determine the license of a
dataset.
- For existing datasets that are re-packaged, both the original license and
the license of the derived asset (if it has changed) should be provided.
- If this information is not available online, the authors are encouraged
to reach out to the asset's creators.
])
+ #claim(
name: [New Assets],
question: [
Are new assets introduced in the paper well documented and is the
documentation provided alongside the assets?
],
answer: TODO, // Replace by answerYes, answerNo, or answerNA.
justification: TODO,
guidelines: [
- The answer NA means that the paper does not release new assets.
- Researchers should communicate the details of the dataset/code/model as
part of their submissions via structured templates. This includes details
about training, license, limitations, etc.
- The paper should discuss whether and how consent was obtained from people
whose asset is used.
- At submission time, remember to anonymize your assets (if applicable).
You can either create an anonymized URL or include an anonymized zip file.
])
+ #claim(
name: [Crowdsourcing and Research with Human Subjects],
question: [
For crowdsourcing experiments and research with human subjects, does the
paper include the full text of instructions given to participants and
screenshots, if applicable, as well as details about compensation (if any)?
],
answer: TODO, // Replace by answerYes, answerNo, or answerNA.
justification: TODO,
guidelines: [
- The answer NA means that the paper does not involve crowdsourcing nor
research with human subjects.
- Including this information in the supplemental material is fine, but if
the main contribution of the paper involves human subjects, then as much
detail as possible should be included in the main paper.
- According to the NeurIPS Code of Ethics, workers involved in data
collection, curation, or other labor should be paid at least the minimum
wage in the country of the data collector.
])
+ #claim(
name: [
Institutional Review Board (IRB) Approvals or Equivalent for Research with
Human Subjects
],
question: [
Does the paper describe potential risks incurred by study participants,
whether such risks were disclosed to the subjects, and whether
Institutional Review Board (IRB) approvals (or an equivalent
approval/review based on the requirements of your country or institution)
were obtained?
],
answer: TODO, // Replace by answerYes, answerNo, or answerNA.
justification: TODO,
guidelines: [
- The answer NA means that the paper does not involve crowdsourcing nor
research with human subjects.
- Depending on the country in which research is conducted, IRB approval (or
equivalent) may be required for any human subjects research. If you
obtained IRB approval, you should clearly state this in the paper.
- We recognize that the procedures for this may vary significantly between
institutions and locations, and we expect authors to adhere to the
NeurIPS Code of Ethics and the guidelines for their institution.
- For initial submissions, do not include any information that would break
anonymity (if applicable), such as the institution conducting the review.
])

18
2024/documents/by-name/vision-whitepaper/logo.typ Normal file
View file

@ -0,0 +1,18 @@
#let kern(length) = h(length, weak: true)
#let TeX = style(styles => {
let e = measure(text("E"), styles)
let T = "T"
let E = text(baseline: e.height / 2, "E")
let X = "X"
box(T + kern(-0.1667em) + E + kern(-0.125em) + X)
})
#let LaTeX = style(styles => {
let l = measure(text(10pt, "L"), styles)
let a = measure(text(7pt, "A"), styles)
let L = "L"
let A = text(7pt, baseline: a.height - l.height, "A")
box(L + kern(-0.36em) + A + kern(-0.15em) + TeX)
})
#let LaTeXe = style(styles => {
box(LaTeX + sym.space.sixth + [2#text(baseline: 0.3em, $epsilon$)])
})

25
2024/documents/by-name/vision-whitepaper/main.bib Normal file
View file

@ -0,0 +1,25 @@
@article{alexander1994template,
title={Template-based algorithms for connectionist rule extraction},
author={Alexander, Jay and Mozer, Michael C},
journal={Advances in neural information processing systems},
volume={7},
year={1994}
}
@book{bower2012book,
title={The book of GENESIS: exploring realistic neural models with the GEneral NEural SImulation System},
author={Bower, James M and Beeman, David},
year={2012},
publisher={Springer Science \& Business Media}
}
@article{hasselmo1995dynamics,
title={Dynamics of learning and recall at excitatory recurrent synapses and cholinergic modulation in rat hippocampal region CA3},
author={Hasselmo, Michael E and Schnell, Eric and Barkai, Edi},
journal={Journal of Neuroscience},
volume={15},
number={7},
pages={5249--5262},
year={1995},
publisher={Soc Neuroscience}
}

99
2024/documents/by-name/vision-whitepaper/main.typ Normal file
View file

@ -0,0 +1,99 @@
#import "@preview/bloated-neurips:0.5.1": (
botrule,
midrule,
neurips2024,
paragraph,
toprule,
url,
)
#import "./logo.typ": LaTeX, LaTeXe, TeX
#let affls = (
ucsb: (
// department: "AI Center",
institution: "University of California, Santa Barbara",
country: "United States",
),
)
#let authors = (
(
name: "Youwen Wu",
affl: "ucsb",
email: "youwen@ucsb.edu",
equal: true,
),
)
#show: neurips2024.with(
title: [Towards More Accessible Scientific Infrastructure: A Neural Vision Pipeline to Interface with Experiments],
authors: (authors, affls),
keywords: ("Machine Learning", "NeurIPS"),
abstract: [
Scientific instruments are often designed to be operated by humans. As
such, they are outfitted with analog dials and controls which are difficult
for machines to understand. In order to ameliorate the inaccessibility of
experimental equipment in fundamental disciplines such as quantum physics,
we seek a systematic approach to processing existing _analog systems_ into
_digital data_ without invasively augmenting them with sensors. In this
paper, we explore the state of the art in computer vision and their
applications in analyzing experimental instruments through a purely vision
based approach. We train a convolutional neural network to triangulate
visual fiducials and construct a pipeline to apply perspective warp based
corrections to normalize images of measurements. We end by designing
_Dendrite_, an end-to-end vision pipeline that can obtain detailed
digital readings from a video stream of an analog instrument.
],
bibliography: bibliography("main.bib"),
bibliography-opts: (title: none, full: true), // Only for example paper.
appendix: [
#include "appendix.typ"
#include "checklist.typ"
],
accepted: true,
)
= Introduction
The rise of online resources in scientific pedagogy has become increasingly
prevalent. Around the world, students use virtual labs that simulate physical
phenomena. However, still lacking is the accessibility of real world hardware
to obtain real results. Experimental instruments are expensive and difficult to
justify for many schools and institutions. One solution to this problem is to
provide shared equipment that is accessible and controlled over the internet.
This allows equipment located in a single place to be used from anywhere in the
world.
One way to build these systems is to augment existing devices with the
capability to be controlled over the internet. However, many scientific
instruments are designed with human operation in mind and contain many analog
dials, readouts, and controls. We seek a way to non-invasively digitize these
devices. Here non-invasively means that we should not perform any irreversible
or drastic changes to the hardware. Digitize refers to obtaining all relevant
outputs as digital data that can be processed by computers, and being able to
operate relevant controls over digital protocols (such as the internet). In
this paper, we focus primarily on obtaining the outputs.
We propose a system which uses an end-to-end vision pipeline that can scan
readouts and translate them into data. Then, the data can be streamed to
virtual simulations which will react exactly as the real life equipment does.
== Requirements
Our end-to-end pipeline will consist of a component to locate the desired
instrument in the image and determine the corrections needed to transform the
image into a point of view where it is directly visible. This may be a neural
network based model that identifies a key fiducial from which we can
extrapolate the perspective transforms needed to bring the image to a
normalized state (here normalized refers to a flattened 2D image that can be
easily analyzed by computer vision).
We then extrapolate from that data to map out all of the various points of
interest. From that point, we can run specialized models on readouts such as
dials to determine their readings.
= The state of the art
We first

35
2024/documents/by-name/vision-whitepaper/package.nix Normal file
View file

@ -0,0 +1,35 @@
{
pkgs,
typstPackagesCache,
typixLib,
cleanTypstSource,
...
}:
let
src = cleanTypstSource ./.;
commonArgs = {
typstSource = "main.typ";
fontPaths = [
# Add paths to fonts here
# "${pkgs.roboto}/share/fonts/truetype"
];
virtualPaths = [
# Add paths that must be locally accessible to typst here
# {
# dest = "icons";
# src = "${inputs.font-awesome}/svgs/regular";
# }
];
XDG_CACHE_HOME = typstPackagesCache;
};
in
typixLib.buildTypstProject (
commonArgs
// {
inherit src;
}
)