I'm also interested in papers from many disciplines, so the wider the range of domains, the higher the value!
1. It's quite readable as a narrative.
2. The maths is not pages of first principle derivations as if the reader is not familiar with the basics of algebraic substitution.
3. The diagrams and graphs are genuinely useful and remove the need for many, many thousands of words that others may have used instead of, or in addition to, the core narrative.
4. It deals with an abstract concept but roots it in concrete mathematical and physical terms. He touches on specific examples.
5. It's quite short given the breadth of subject area.
[1] https://people.math.harvard.edu/~ctm/home/text/others/shanno...
This is the paper I was going to cite as well.
https://en.m.wikipedia.org/wiki/Communication_Theory_of_Secr...
IMO Shannon does an amazing job laying out the foundational ideas behind cryptographic security in a way that is very accessible to anyone with a basic understanding of probability.
The clarity and simplicity that he achieves is especially striking when you compare his approach to what you’ll see from modern cryptographers. No offense to those guys - they’re amazing - but accessible their stuff is not.
In a system with growing inequality where the rich benefit at the expense of the poor, this artificial redistribution can go on for some time, but once the inequality gets so bad that people revolt, then the amount of "guard labor" that needs to be performed goes up. Poverty and desperation makes people more likely to perform "guard labor" because it gives them a chance to escape poverty and avoid being targeted themselves which further feeds into authoritarian politicians gaining more power as they have no trouble finding soldiers willing to maintain the inequality. This works but only until guard labor reaches such a critical mass that half the population engages in it. Once that point is crossed, guard labor will start defecting against the current political leadership and conduct a military coup.
Presents a beautiful algorithm (maximum matching in general graphs) and is very well written
Often called "the paper which saved the internet" due to solving congestion collapse on the ARPANET, and inventing the fundamentals of TCP Congestion Control still used countless times every single day on all computers everywhere. It's very readable and presents complex math in easily understood graphs for non-math people.
[0] https://www.cs.cmu.edu/~rdriley/487/papers/Thompson_1984_Ref...
Even if you argue that Bitcoin as it has evolved has flaws, it is the best implementation we’ve come up with so far. Like democracy, it isn’t perfect but it is much better for its use case than anything else we have come up with to date.
You miss something only when it's gone.
The abstract begins with "It is easy to argue that real signals must be bandlimited. It is also easy to argue they cannot be so. This paper presents one possible resolution of this paradox"
http://web.eng.ucsd.edu/~massimo/ECE287C/Handouts_files/On-B...
The best leadership paper I've read.
Who Are They to Judge?: Overcoming Anthropometry and a Framework for Fat Bodybuilding [1]
Separately, Alan Sokal has written a measured take on Robin DiAngelo's best selling book on "White Fragility"[2], which thrived on the mania post-George Floyd.
[0] https://areomagazine.com/2018/10/02/academic-grievance-studi...
[1] https://www.tandfonline.com/doi/pdf/10.1080/21604851.2018.14...
[2] https://physics.nyu.edu/sokal/white_fragility_FINAL2.pdf
Can you give some context for this that isn't a 47 minute read? What might someone think it does or does not demonstrate?
https://rust-class.org/static/classes/class19/dijkstra.pdf
The entire paper is just one side of one page. In it he describes a basic, fundamental concurrency problem, gives a solution, and then provides a proof of the solution.
Nothing else has ever made me appreciate how hard science really is and how little the general public understands it.
Presents a really nice conceptual/intuitive explanation of how/why it works, rather than the traditional algebra based definition/proof you get in many linear algebra courses.
[1] https://www.wiley.com/en-us/The+Warming+Papers%3A+The+Scient...
Related: You might find the book Matters Computational - https://www.jjj.de/fxt/#fxtbook and the FXT library it describes - https://www.jjj.de/fxt/ useful.
https://arxiv.org/pdf/1410.0369.pdf
Possible Girls:
Ironically, I think the paper presents more than just the idea and examples of statecharts, rather it also _implicitly_ contains a _method_ for discovering mechanism - the long winded example of the author's digital watch, in my eyes, is a marvel.
For C++ I used Kris Jusiak's library [boost].sml extensively. He too has a few talks on the subject, in particular his c++now 2017 talk.
I came to know of Statecharts (from UML) long ago, but had not bothered to study it properly, thinking it was just a "more complicated State Machine representation". Seems i was wrong and there is much to learn here - https://statecharts.dev/
I read this a few years back as I was going down an object-capability rabbit hole and found it extremely compelling. (And also made me disappointed that most of the systems we use today do not work like this! Code execution vulnerabilities would be so much less immediately hazardous if they did.)
https://www.frontiersin.org/articles/10.3389/fpsyg.2015.0110...
You’ll Never Guess Who Wrote That: 78 Surprising Authors of Psychological Publications
https://scottlilienfeld.com/wp-content/uploads/2021/01/lilie...
Interest as the Missing Motivator in Self-Regulation
https://www.researchgate.net/profile/Dustin-Thoman/publicati...
“The Spandrels of San Marco and the Panglossian Paradigm”, (Gould et al)
“ A quantitative description of membrane current and its application to conduction and excitation in nerve” (hodgkin and huxley)
a few other that don’t come to mind right now
However thus far, a paper that literally changed my life: "Value Dependence Graphs: Representation Without Taxation", D. Weise, R. F. Crew, M. Ernst, B. Steensgaard, POPL 1994. (This was the proverbial butterfly flap that moved me through three countries).
There are many many other good papers and it's not a one-dimension metric so it's hard to pick out winners.
This paper presents a cure for an extremely aggressive cancer using vitamin A and arsenic. Its a unique, relatively benign treatment strategy that completely avoids chemotherapy. As far as I know this is the best result in all of oncology, though the cancer it treats is very rare.
The most well known paper in oncology that is probably more interesting to a general audience is "The Hallmarks of Cancer" by Hanahan and Weinberg.[1]
[0] https://www.nejm.org/doi/full/10.1056/nejmoa1300874
[1] https://www.cell.com/cell/fulltext/S0092-8674(00)81683-9?_re...
https://www.cs.kent.ac.uk/people/staff/dat/miranda/whyfp90.p...
Life at Low Reynolds Number makes you think of things which arent visible but move :https://www.damtp.cam.ac.uk/user/gold/pdfs/purcell.pdf
There are more but lets see if the above two work for you first
https://dl.acm.org/doi/abs/10.1145/360018.360022
Their work on list processing was an inspiration for John McCarthy's famous paper on Lisp:
That should give you dozens of recommendations and a bit of rationale for them.
It's hard to say what best means, but some papers play their part better than others. Not every paper should change the field, nor should they all be the length of a novel.
The Matrix Mechanism (2015) https://people.cs.umass.edu/~miklau/assets/pubs/dp/Li15matri... Good introduction to the select-measure-reconstruct approach to answering large sets of linear queries (e.g. histograms, averages, ...). This paper is important for the statistics approach to DP synthetic data (which is what I primarily work on): https://differentialprivacy.org/synth-data-1/
Differentially Private Quantiles (2021) https://arxiv.org/pdf/2102.08244.pdf It turns out that medians are difficult to deal with under DP. This paper introduces a method called JointExp to estimate several quantiles at once. One innovation here is that the user doesn't have to know anything about the distribution of the variable in advance for it to work.
Boundary Differential Privacy (2020) https://www.computer.org/csdl/journal/tq/5555/01/09286504/1p... The main motivation for DP is to protect the privacy of people whose data is used to compute some statistic. This paper, instead, uses DP techniques as a way to inject controlled noise into the output distribution of a classifier to defend against model reconstruction attacks.
Also this one: "The Internet (Never) Forgets (2017) PDF" https://scholar.smu.edu/cgi/viewcontent.cgi?article=1037&con...
End to end principle: https://web.mit.edu/Saltzer/www/publications/endtoend/endtoe...
The rise of "worse is better": https://www.jwz.org/doc/worse-is-better.html
It is very clearly written does not contain any maths and can be understood by a non expert (IMO) definitely worth a read.
The firs one form 2010 tries to find using multi agents simulation the best approach to solve the Peter Principle in organizations (If you do not know what is the Peter principle is the reason why your boss is so incompetent but still earns more than you)
https://www.sciencedirect.com/science/article/pii/S037843710...
the second is from the same author a work (ignobel 2022) that demonstrate that in the long run it does not matter how prepared and skilled you are.Your income depends more from the fortune and chances that you had in your life than on how prepared you are.
Absolutely fantastic paper summarizing the field of automata in the 1950s. Amazing how amazing of a surveyor Shannon was. I wish there was a present-day analog for the field now.
These quotes in particular may be of interest to you if you are unsure or limited in time. If they sound interesting, read through the rest of the paper.
“We lecturers naturally worry about the content of our lectures rather than the emotions we express in giving them. As human beings, students respond immediately to the emotive charge, even if they do not understand the content. The lecturer may have tried to give a balanced account of the debate between X and Y, but his preference for Y shines through. When the students come to write the essay on the relative merits of X and Y, they know where to put their money. The lecturer might try to balance the lecture by suppressing his enthusiasm for Y, but this 'objective' presentation will make a mystery of the whole exercise. The students will wonder why they have to sit through all this stuff about X and Y when even the lecturer does not seem to care much for either of them. The better strategy is for the lecturer to plunge into the works of X, reconstruct X's mental world and re-enact X's thoughts until he shares some of X's intellectual passions. We can be sure that X had intellectual passions, else we would not now have the works of X.”
“Remember Dr. Richards on Tolstoy? If Tolstoy's view really is 'plainly untrue then there is no point mentioning it. If Tolstoy's view is worth mentioning, then it is worth inhabiting Tolstoy's position, reconstructing his thought and thus feeling the force of his motives. Nothing less will bring Tolstoy's thought to life, which you have to do if the students are to see any point in learning about it. Victory over a corpse is no less pyrrhic than victory over a straw man.”
https://uhra.herts.ac.uk/bitstream/handle/2299/5831/903260.p...
Their methodology is brilliant. I want this research replicated, all over the place, all the time, with different variables.
The work provided an elegant and tractable way to model complex chemisry and a lot of physics. Most of molecular level modeling that cuts across electronics, energy and pharma are based on this work.
[1] https://journals.aps.org/pr/abstract/10.1103/PhysRev.136.B86...
Have you seen anything worth reading in that line of literature which addresses a more practical issue of segment sizing and segment overlap on accuracy?
Humans display a reduced set of consistent behavioral phenotypes in dyadic games: https://www.science.org/doi/10.1126/sciadv.1600451
Simple mathematical models with very complicated dynamics: https://www.researchgate.net/publication/237005499_Simple_Ma...
It's a paradigm shift. Entropy (in the Bayesian sense) is fundamentally a subjective quantity. Maximizing entropy is just minimizing the assumed information. When dealing with large complex systems over long timescales, the only pieces of information one is justified in assuming are the values of those quantities that are conserved globally by the dynamics (eg, total energy, total mass, etc). The "thermodynamic entropy" S is just the maximum entropy, given a certain set of conserved quantities (an "ensemble") -- it is therefore more or less objective, modulo the conserved quantities.
The second law of thermodynamics is just the information processing inequality: you don't have any more information about the future state of the system than you do about the present state. If anything, you have less, because if you have any information about the current state of non-conserved quantities, that information is not valid at other times (assuming you don't have complete information & the ability to fully simulate the dynamics). From this perspective, entropy does not generate the "arrow of time": the argument is symmetric in time.
Another from Jaynes worthy of mention: "Prior probabilities" [1] discusses the use of group theory to derive "non-informative" prior distributions by considering the set of transformations that result in equivalent inference scenarios. This is resolved another question I had when studying stat mech: "How come we start with an assumption of uniform probability in phase space as expressed in (x,p) coordinates -- if we transformed to a different coordinate system, it would look like a non-uniform distribution!" The answer is that we assume Galilean invariance. The concept applies outside stat mech as well.
Plenty of other interesting stuff [2].
[0] https://bayes.wustl.edu/etj/articles/theory.1.pdf [1] https://bayes.wustl.edu/etj/articles/prior.pdf [2] https://bayes.wustl.edu/etj/node1.html
by Jim Gray
His wikipedia page lists a few publications that could serve as starting points:
https://en.wikipedia.org/wiki/Charles_J._Fillmore#Publicatio...
(Some of my favourites in that area used to be Martin Haspelmath, Balthasar Bickel, Gilbert Lazard, ... - but I haven't kept myself up to date with the literature for years now)
This classic in the field of biogeochemistry is so well written --its both technical and accessible, digging into a fundamental puzzle that underlies so many patterns in life on earth.
On the other end of the spectrum is "Anyons in an exactly solved model and beyond" by Kitaev. It's very long, and yet densily filled with ideas. I'm getting back to it after years, and still finding out new things.
A footnote in scientific research on folk, this two-page-worth of text is a rebuttal to the scientific journal equivalent to a Hacker News comment in which the authors come close to very nearly questioning the validity of someone's PhD.
Just goes to show: No matter what field you're in, someone is going to incorrect you.
What did I like about these papers? The content I guess. :p Come to think of it, the accompanying website is good too. Much better than just the paper!
Is the paper that has influenced me the most.
Original IPFS paper is one of worst papers that I had read.
“On the Common Law Origin of the Infield Fly Rule”
A rare blend of honest academic inquiry and parody writing
It is about traditional martial arts masters, trapped in their echo chamber, sniffing their own farts. The whole industry gets its ass kicked by mixed martial arts. Basically street thugs versus shaolin kung fu masters.
it describes in-group bias, echo chambers, and cognitive dissonance in large groups. Very applicable in modern science, politics and so on.
While i agree that many treat "Traditional Martial Arts" almost like a "Religion" (i.e. The Ancients knew everything and thou shalt not question) we also need to be careful that we don't throw the "baby out with the bathwater". Almost all of current-day martial arts are "Sports" (including MMA etc.) with rules and refereeing in place to avoid serious injury/death. Thus while we have a lot of modern knowledge w.r.t. Anatomy/Physiology etc. much of the mental training needed for "life-or-death" situations have been lost. A technique by itself is nothing; it is the "killing mentality" needed to go with it that is everything. I think that is what we need to refocus on from "Traditional Martial Arts" while at the same time interpreting them in the light of Modern Science/Psychology.
Specialized people (military, self defense contractors, etc.) might optimize for lethality. Maybe Krav Maga has a greater focus on that than most. But most of the famous Eastern arts aren't delaying with life and death stuff anymore either. It's not Kill Bill.
The paper bemoans the BS in "Traditional Martial Arts" but misses the point that they are no longer taught with the same level of intensity/seriousness/viciousness/ruthlessness that it was originally developed with. Lacking this training mindset the "Martial" art becomes merely a "Sport". But many of the ancient texts cover these principles quite well and are worth learning and interpreting in modern physiological/psychological terms.
Imagine a hypothetical new martial arts that allows all the moves of MMA, but with biting and eye poke allowed, so practitioners of this arts is at a definite advantage against MMA. Would you say that this martial art is any better?
In this context, some martial arts are dramatically better than others.
MMA + Eye-poking martial art: Yes.
So am I advocating scoffing at the word of your sensei or senpai? No. That’s not being an epistemically responsible agent, that’s just being an asshole. All the old constraints on your behaviour still apply. I’m arguing for the importance of being cautious in what you believe.
Instead of saying they are "trapped in their echo chamber, sniffing their own farts". Although I guess you inevitably sniff a few things you wouldn't normally chose to in a dojo (such as other peoples' sweat or foot smell).
Interestingly the author then goes on to quote "the words of the Buddha "copied from their place of honour on the wall of Harry Cook’s dojo":
Do not believe on the strength of traditions even if they have been held in honour for many generations and in many places; do not believe anything because many people speak of it; do not believe on the strength of sages of old times; do not believe that which you have yourselves imagined, thinking that a god has inspired you. Believe nothing which depends only on the authority of your masters or priests. After investigation, believe that which you have yourselves tested and found reasonable, and which is for your good and that of others.
And while that sounds very enlightened to the modern reader, there's at least this guy who thinks it's a mistranslation adjusted to that modern reader's sensibilities:
https://fakebuddhaquotes.com/do-not-believe-in-anything-simp...
Anyway an interesting read, so thanks for linking.