Machine Learning and Complexity

 

Schloss Dagstuhl by Monet by Dall-E

At Dagstuhl earlier this month, I hung out for a little bit with the participants of the other seminar, Knowledge Graphs and their Role in the Knowledge Engineering of the 21st Century. Knowledge graphs are what you would expect them to be, nodes are objects like "Berlin" and "Germany" with directed edges with labels like "capital". Think of having knowledge graphs of hundreds of millions of nodes and how that could help answer queries about the world. These secondary workshops are shorter and focus on creating a new vision, in this case how to maximize the importance of knowledge graphs in an increasing ML-focused world.

Perhaps we need such a visioning seminar for complexity. While we often get lost in the mathematical questions and techniques in our field, computational complexity is designed to understand the difficulty of solving various problems. Machine learning and advances in optimization should be changing that conversation. If you imagine a world where P = NP (and I did exactly that in chapter 2 of my 2013 book) much of what you consider is starting to happen anyway. ML does fail to break cryptography but then again, isn't this the best of all possible worlds? 

Look at what Scott Aaronson said back in 2006.

If P=NP, then the world would be a profoundly different place than we usually assume it to be. There would be no special value in “creative leaps,” no fundamental gap between solving a problem and recognizing the solution once it’s found. Everyone who could appreciate a symphony would be Mozart; everyone who could follow a step-by-step argument would be Gauss; everyone who could recognize a good investment strategy would be Warren Buffett. 

If I can be a Monet, can Mozart be far behind? ML trading by some hedge funds are beating Warren Buffett but remember if everyone trades perfectly, no one beats the average. Gauss is going to be trickier but it's coming. There's a reason Scott is spending a year at OpenAI to understand "what, if anything, can computational complexity contribute to a principled understanding of how to get an AI to do what we want and not do what we don’t want".

Is the complexity of approximating Vertex Cover of degree 3 open? (ADDED LATER-NO)

 RECALL:

A max-problem f has a A P Time App Scheme (PTAS) if there is an algorithm ALG such that

 ALG(\epsilon) \ge (1-\epsilon)f(x).

A min-problem f has a A P Time App Scheme (PTAS) if there is an algorithm ALG such that

 ALG(\epsilon) \le (1+\epsilon)f(x).

(Note that the poly can depend on epsilon so it may be something like n^{1/epsilon}.)

MAX3SAT is, given a formula with \le 3 literals per clause, find an assignment

that maximized the number of clauses satisfied.

VCB-a is Vertex cover where graphs have degree \le a

The following are known:

0) MAX3SAT is in APX.

1) The PCP paper, here, showed that if MAX3SAT has a PTAS then P=NP.

2) Papadimitriou and Yannakakis (here)  had showed much earlier that MAX3SAT \le VCB-4 with an approx preserving reduction.

3) From (1) and (2) we have that VCB-4 has a PTAS then P=NP. (VC is in APX by an easy 2-approx).

4) Clearly VCB-2 is in P.

The following seems to be open, though if you know otherwise pleae leave a comment:

Is VCB-3 a) in P? b) NPC? (ADDED LATER- NPC- See comments.) 

Is the following true: if VCB-3 has a PTAS then P=NP. (ADDED LATER- NO PTAS-See Comments)

NOTE- all of the above is true for Ind Set-4 and Dom Set-4. So that leads to more open problems.

POSTED UPDATED VERSION OF Computers and Intractability: A guide to Algorithmic Lower Bounds posted (New title)

We have posted a revised version of 

Computational Intractability: A Guide to Algorithmic Lower Bounds

by Demaine-Gasarch-Hajiaghayi

The book is here.

(For the original post about it, edited it to use the new title (see below), see HERE.) 

We  changed the title (the title above is the new one) 

since the earlier title looked too much

like the title of Garey's and Johnson's classic. While that was intentional we 

later felt that it was too close to their title and might cause confusion. 

Of course changing the title might also cause confusion; however, 

this post (and we will email various people as well) will stem that confusion. 

We welcome corrections, suggestions and comments on the book. Email us at hardness-book@mit.edu

There are two different definitions of Integer Programming. Why?

Alice and Bob have the following conversation.

===============================

ALICE: In your book you define INT PROG as, given a matrix A and vectors b,c,

find the integer vector x such that Ax\le b and c DOT x is maximized.

This is not correct! You also need x\ge 0.

BOB: Really? I always heard it without that extra constraint, though I am

sure they are equivalent and both NP-complete (Alice nods).

Where did you see it defined with that extra constraint?

ALICE:

Wikipedia entry in IP

Chapter of a book at an MIT website

Something on Science Direct

A course at Duke

An article by Papadimitriou 

An article on arxiv

The book Graphs, Networks and Algorithms by Dieter Jungnickel

Bob, do you have examples where they do not use that extra constraint. 

BOB: 

Math Works

Lecture notes from UIUC

Lecture notes from Lehigh Univ.

The book Parameterized Complexity Theory by Flum and Grohe

The book Computers and Intractability : A Guide to the Theory of NP-Completeness by Garey and Johnson

ALICE: Both of our lists are impressive. So now what? 

--------------------------------------------------------------------

(This is Bill again.)

What indeed!

1) Why are there two definitions of Int Prog?

2) When is it good to use which one? 

Monarachy: A Problem with Definitions

 As I am sure you know, Queen Elizabeth II passed away at the age of 96 recently.  I am not a royal-watcher, but I am a royal-watcher-watcher. That is, the question of why people care about the lives of these people intrigues me. A few notes

1) Was she a good Queen? People tend to think so; however, since the job is somewhat ill-defined its hard to say. 

2) The Queen is supposed to be above politics (she does not vote- I was surprised to find out that legally she can, but she really can't). We know very few of Queen Elizabeth II's opinions on political events. But the notion of political is not well defined. One would think that if she did an appeal for people to take the COVID vax that would not be political, but somehow it is (I do not know if she did such an appeal). King Charles III believes in global warming and that we need to do something about it. This again should not be political but is. 

3) She is the second longest reigning Monarch. First is King Louis XIV who first became king at the age of 4. I had a blog complaining about this here. However, there is a more interesting point I want to make. From the first to the last day of King Louis XIV reign not much had changed. Technology, politics, other things just didn't change much. By contrast the world changed A LOT between Queen Elizabeth II first and last day:

a) The British were an important power in 1952. Less so now.

b) When her father died she was in Kenya and it took 4 hours to get the news to her. Now that would be immediate. 

c) Divorce was considered bad in 1952 and is why King Edmond VIII could not be king (he wanted to marry a twice-divorced woman whose ex-husbands were still alive). And now three of the Queen's children have been divorced.

d) Gay people.. enough said. There has even been a royal gay wedding, see here

Black people (can't call them African-Americans), Women,... you fill it in. 

e) When Charles wanted to get married it seemed to be important that he marry a virgin. We cannot imagine this mentality anymore. When Prince William and Kate got married they were already living together and this was NOT an issue for ANYONE. I looked up what the Church of England thought of it and all I got was some very bland comments like That's what young people do nowadays. 

3) Is the monarchy a good thing? As an American I feel I do not have a right to an opinion. If the citizens of the United Kingdom approve of the monarch (polls show they do) then who am I do tell them they are wrong? Even so, lets look at reasons for it

a) Tourism. It has been said that the Monarchy leads to MONEY from tourism. So it is worth the price? Nobody seems to know and it would be hard to tell. However, I don't think the citizens of the United Kingdom view  money as the reason for Monarchy. The American analog is giving Disneyland tax breaks to be in Florida which generates jobs. I doubt they think of the Monarchy in those mundane transactional terms. 

b) CS Lewis said 

Where men are forbidden to honour a king they honour millionaires, athletes, or film stars instead: even famous prostitutes and gangsters. For spiritual nature, like bodily nature, will be served; deny it food and it will gobble poison.

This is  bit odd- they must all pretend to like the monarchy to make it work. A long time ago when Charles and Dianna were both having affairs, 80% of the citizens the United Kingdom thought that was okay so long as they are discreet so the people don't find out. But- those ARE the people.

Also odd- CS Lewis was a theologian and a  believing Christian; however, his comment above can apply to God as well as to Kings. 

Thirty Years of Dagstuhl

 

Dagstuhl old-timers at the original castle

I'm back at Dagstuhl for the seminar on Algebraic and Analytic Methods in Computational Complexity. My first seminar at Dagstuhl was back in 1992. I've been coming for thirty years and have been here roughly thirty times. My last trip was pre-covid (virtual Dagstuhls don't count) and I really needed this chance to hang out and talk complexity with colleagues old and new.

Some changes since my last trip. The room doors have locks (there are rumors of an incident). You have to create your own keycard on a new machine logging into your Dagstuhl account. I had a long random password through a password manager and it was not so easy as process.

The main conference room has been updated with tech for hybrid meetings, and new led lights. Books were removed from the library to create a coffee breakout space.

No Bill this time so no typecasts. Still the best part of the week is talking and hearing about complexity. Today I learned about the orientations of Sperner's lemma, that there is one more triangle oriented according to the direction of the corner vertices than those oriented the other way. Christian Ikenmeyer used this fact to motivate a study of closure properties of #P-functions.

The NIST Process for Post-Quantum Cryptography

Guest post by Jonathan Katz

Over the past few months there have been several interesting developments in the NIST post-quantum standardization process.

By way of background, since the advent of Shor's algorithm in 1994 we have known that a large-scale, general-purpose quantum computer would be able to break all currently deployed public-key cryptography in (quantum) polynomial time. While estimates vary as to when (or even whether!) quantum computers will become a realistic threat to existing public-key cryptosystems, it seems prudent to already begin developing/deploying newer "post-quantum" schemes that are plausibly secure against quantum computers.

With the above in mind, NIST initiated an open process in 2017 for designing post-quantum cryptographic standards. Researchers from around the world submitted candidate algorithms for public-key encryption/key exchange and digital signatures. These were winnowed down over a series of rounds as cryptographers publicly debated the relative merits of different proposals, or showed security weaknesses in some candidates.

On July 5 of this year, NIST announced that it had selected four of the submissions as finalists for standardization. Only one candidate for public-key encryption was chosen, along with three digital signature schemes. Three of the four selected algorithms rely on the hardness of lattice problems; the only non-lattice scheme is a hash-based signature scheme. (It is possible to build digital signatures using "symmetric-key" assumptions alone.) In addition, four other public-key encryption schemes not based on lattices were designated for further study and possible standardization at a later point in time.

Less than one month later, Castryck and Decru announced a classical attack on SIKE, one of the public-key encryption schemes chosen for further study. SIKE is based on a conjectured hard problem related to isogenies on supersingular elliptic curves. The attack was not just theoretical; the researchers were able to implement the attack and run it in less than a day or less, depending on the security level being considered. Details of the attack are quite complex, but Galbraith gives a high-level overview. Subsequent improvements to the attack followed.

It is worth adding that the above follows an entirely classical attack shown roughly six months earlier on Rainbow, another submission to the NIST standardization process that made it to the 3rd round. (Rainbow is a signature scheme that relies on an entirely different mathematical problem than SIKE.) For completeness, note that none of the four finalists are impacted by any of these attacks.

A few reflections on the above:

• It is amazing that the factoring and RSA problems are still hard (for classical computers), more than 40 used after they were proposed for cryptography. The same goes for the discrete-logarithm problem (in certain groups).
• It is not easy to find other hard mathematical problems! Code-based cryptography has been around about as long as factoring, but has been somewhat unpopular for reasons of efficiency. Lattice-based cryptosystems still seem to give the leading candidates.
• We need more (non-cryptographers) studying cryptographic assumptions. The attacks on SIKE involved deep mathematics; attacks on lattice problems may involve algorithmic ideas that cryptographers haven't thought of.

Computational Intractability: A Guide to Algorithmic Lower Bounds. First draft available! Comments Welcome!

(This post is written by Erik Demaine, William Gasarch, and Mohammad Hajiaghayi)

In 1979 Garey and Johnson published the classic

        Computers and  Intractability: A Guide to NP-Completeness

There has been A LOT of work on lower bounds since then.

Topics that were unknown in 1979 include

Parameterized Complexity,

 Lower bounds on approximation,

Other hardness assumptions (ETH, 3SUM-conjecture, APSP-conjecture, UGC, Others), 

Online Algorithms,

Streaming Algorithms, 

Polynomial Parity Arguments, 

Parallelism, and 

Many new problems have been shown complete or hard in NP, PSPACE, and other classes.

Hence a sequel is needed. While it is impossible for one book to encompass all, or even a large fraction, of the work since then, we are proud to announce a book that covers some of that material:

Computational Intractability: A Guide to Algorithmic Lower Bounds

by Erik Demaine, William Gasarch, and Mohammad Hajiaghayi. MIT Press. 2024

See HERE for a link to a first draft.

We welcome corrections, suggestions and comments on the book. Either leave a comment on this blog post or emailing us at hardness-book@mit.edu

20 Years of the Computational Complexity Weblog

I first posted on this blog twenty years ago today, still the oldest and longest running weblog in theoretical computer science, possibly in all of computer science. In those twenty years we've had nearly 3000 posts and over 23,000 comments and 10,000,000 page views. Bill Gasarch joined me officially as a co-blogger over 15 years ago on March 30, 2007. 

We've seen major results in computational complexity but the biggest challenges remain, in particular major separations of complexity classes. We've also had a front row seat to a computing revolution with  the growth of cloud and mobile computing, social networks connecting us for better or for worse, and incredible successes of machine learning. It's almost as though we've been handed an oracle that gives us much of the goodness of P = NP while leaving cryptography intact. 

What will the next twenty years bring? We'll be watching, posting and tweeting. Hope you keep reading and commenting. 

Conference Modality

We have had an almost normal summer conference season, for some sense of normal. At one of those conferences I participated in an hybrid conversation about whether the conference should be in-person, virtual or hybrid the following year. Here are some general thoughts.

In-Person

The traditional conference format. People travel from near and far to a hotel, conference center or campus location. Talks given in large rooms, often in parallel. A reception, some social activities, participants gathering in small groups to go out for meals. 

Positives: In-person maximizes interaction between participants. Being physically away from your home means you can focus your time on the conference and your fellow participants. This was more true before the mobile devices/laptops/internet days, but still most participants will spend more time on-site than on-line.

Negatives: Expensive--with registration, hotel and air fare, even a domestic participant could have to pay $2000 or up, higher for those traveling internationally. Visas can be hard to get. Some still feel unsafe in large groups. People often leave early, pity the last speakers. And don't forget the carbon footprint. 

As countries declare war on other countries or states deny certain rights, there is a push against meetings in certain places. Note the disclaimer for next year's FCRC. You might upset some people if you have conferences at these locations (and others if you don't).

Virtual

Virtual conferences would never in the past have been taken seriously but Covid forced our hands. 

Talks are streamed or pre-recorded. Some interaction with chats in talks, zoom get togethers or though a systems like virtual chair. Even if we had a perfect "metaverse" experience where we could get together as though we were in person, not being there in person means we wouldn't make it a priority.

The big advantages are costs are low, it's easy to attend talks, and no danger of spreading disease. Still a virtual conference can feel too much like just a collection of streamed and recorded talks.

Hybrid

So let's make the conference hybrid and have the best of both worlds. Alas, it doesn't work out that way. It's nearly impossible to have good interaction between in-person and virtual participants--basically you have to run two separate meetings. Do you allow virtual talks or require an author to show up in person. 

How do you set prices? Lower prices for virtual increases assess but decreases in-person attendance. Participants (or their advisors) might opt to save expenses and attend virtually instead of in-person, reducing the networking opportunities for everyone. 

Most conferences tend to take the hybrid route to avoid the complaints if one went fully in-person or virtual, but hybrid just pretty much guarantees a mediocre experience for all.

Opinion

My suggestion is some years run the conference virtually and others in hybrid. We already have too many conferences, a byproduct of our field using conferences as the primary publication venue. I suggest following conferences like the International Congress of Mathematicians or the Game Theory World Congress, held every four years. If the main conference of a field is held every four years, researchers, particularly senior researchers, make a bigger effort to be there. You can have the virtual meetings the other years so researchers, particularly students, can continue to present their work.

No easy solutions and CS conferences have not worked well for years. Maybe the struggle to define future conferences will allow us to focus more on the connecting researchers than just "journals that meet in a hotel".

A non-controversial question about the Documents Donald Trump had in his house

This is a non-partisan post. In the interest of disclosure I will divulge that I do not think private citizens should have top secret government documents in their house.

One phrase I kept hearing in the reporting was (I paraphrase and may have the number wrong)

                                  The FBI removed 15 boxes of documents

Documents? Like--- on paper? 

a) Are the documents also online someplace? Perhaps they intentionally are not so that they can't be hacked. 

b) Is having top secret documents only on paper safer than having them in electronic form? Normally I would think so. Donald Trump  having them is a very unusual case. 

c) Having to store all of those documents on paper would seem to have storage problems. I can imagine someone with NO bad purposes making copies and taking them home since they are tired of only being about to read them in a special room. 

d) A problem with having them ONLY on paper is that if an accident happens and they get destroyed there is no backup. Or is there? Are there copies somewhere? That leads to twice the storage problems. 

e) There is a tradeoff of security and convenience. Is having the documents only on paper is an extreme point on the tradeoff, but it may be the right one. It may depend on how important it is to keep the documents secret. 

f) I've heard (apocryphal?) stories about some top secret document also available in public though quite legal sources (e.g., a physics journal that discusses nuclear stuff).  Does the government make to much classified? If so then the problem arises of people not taking the classification seriously and getting careless. I doubt that is what happened here. 

g) The question I am most curious about is why did he take them? For most of his other actions his motivations are clear (e.g., he is pushing STOP THE STEAL since he wants to be president). But for this one its not clear. Unfortunately,  I doubt we'll ever find out. Maybe the answer is in some documents either on paper or electronic. 

The Godfather of Complexity

Juris Hartmanis 1928-2022

On Friday, July 29th, I was in the immigration line at an airport in Mexico. My phone rang with Bill Gasarch on the Caller ID but starting vacation I declined the call. The voicemail gave me the sad news that Juris Hartmanis, the person who founded computational complexity and brought me into it passed away earlier that day. I messaged Bill and told him to write an obit and I'd follow with something personal when I returned.

Hartmanis and Stearns in 1963

In November 1962 Hartmanis, working with Richard Stearns at the GE Labs in Schenectady, determined how to use Turing machines to formalize the basic idea to measure resources, like time and memory, as a function of the problem being solved. Their classic Turing-award winning paper On the Computational Complexity of Algorithms, not only gave this formulation but showed that increasing resources increased the problems one can solve. The photo above, from a post celebrating the 50th anniversary of the paper, shows Hartmanis and Stearns with the main theorem of their paper on the board.

Twenty-one years later, a junior at Cornell University still trying to find his way took undergraduate theory from the man himself. Juris brought the topics to life and I found my passion. At the beginning of the class, he said the highest grade usually went to an undergrad followed by the grad students in the class. I was a counterexample, as I had the second highest grade. Never did find out who beat me out.

In spring of my senior year, 1985, I forgave the traditional senior-slump Wines for graduate complexity with Juris. He focused the course around the isomorphism conjecture he developed with his student Len Berman, which implied P≠NP, and Hartmanis believed using the conjecture might lead to settling P v NP. He offered an automatic A to anyone who could prove the isomorphism conjecture. I guess any other proof of P≠NP only warranted a B?

I would later be obsessed by the isomorphism conjecture as an assistant professor, coming up with not one but two oracles making it true. The isomorphism conjecture didn't end up settling P vs NP, but then again neither did any other approach.

It wasn't just me, there was a reason that many of the great American complexity theorists, including Ryan Williams, Scott Aaronson and my own PhD advisor Michael Sipser, were undergrads at Cornell. Many more were PhD students of Hartmanis.

Juris Hartmanis had a certain gravitas in the community. Maybe it was his age, the way he dressed up, his seminal research in the field, or just that Latvian accent. He founded the CS department at Cornell in the 60s and served as head of the CISE directorate at the National Science Foundation in the 90s. His 60th birthday party at the 3rd Structures in Complexity conference (now the Computational Complexity Conference) was the only time I've seen complexity theorists in ties.

Juris Hartmanis (center) being toasted by Janos Simon

A few of my favorite Hartmanis quotes.

• "We all know P is different than NP. We just don't know how to prove it." - Still true.
• "I only make mistakes in the last five minutes of the class." - Sometimes he made a mistake with ten minutes left but only admit it in the last five minutes.
• "Primality is a problem not yet know to be in P but is hanging on by its fingernails with its grip continuing to loosen each day." - Juris Hartmanis said this in 1986, with primality hanging on for another 16 years.

Thanks Juris for creating the foundations of our field and inspiring so many people, yours truly included, to dedicate ourselves to it.

Much more to read:

• In 1981 Hartmanis wrote a personal article for the Annals of History of Computing describing the birth of computational complexity and his role in it.
• A CACM interview with Hartmanis gives a personal story of his path from Latvia to Kansas City to Pasadena to Schenectady to Ithaca and DC.
• Obits by Cornell, Ryan Williams and Dick Lipton and Ken Regan.
• My old blog posts celebrating Hartmanis' 90th birthday and the 50th anniversary of Hartmanis-Stearns,
• And the many retrospectives, remembrances, special issues and conferences to come.

The Held Prize for comb. opt. AND Disc Opt AND Alg AND Complexity theory AND related parts of CS.

 Dan Spielman asked me to blog about the Held Prize. I first present what he send me, and then have some thoughts.

FROM DAN: 

--------------------------------------------------------------------------------------------------

Nominations are now being accepted for the National Academy of Sciences’ 2023 Michael and Sheila Held Prize. The Held Prize honors outstanding, innovative, creative, and influential research in the areas of combinatorial and discrete optimization, or related parts of computer science, such as the design and analysis of algorithms and complexity theory. This $100,000 prize is intended to recognize recent work (defined as published within the last eight years). Additional information, including past recipients, eligibility requirements, and more, can be found at here.

All nominations must be submitted online. Unless otherwise stated, the following materials must be submitted: 

A letter from the nominator describing the candidate's work and why he or she should be selected for the award. No more than three (3) pages.

Curriculum vitae. No more than two (2) pages (similar to CVs included with NSF proposals).

Bibliography listing no more than twelve (12) of the nominee's most significant publications.

Suggested citation. A 50-word summary stating why the nominee should be considered for this award. (Citation

examples)

Two letters of support. Support letters must be written by individuals from institutions outside both the

nominator's and the nominee’s institution. Up to three letters of support are accepted.

Nominations will be accepted through Monday, October 3, 2022. Please help spread the word that the nomination process is underway. 

----------------------------------------------------------------------------------------------------

BILL COMMENTS

1) The scope seems rather broad (Dan confirmed this in private email) in that its Comb Opt AND Discrete Opt OR related fields like algorithms and complexity theory. 

2) The research has to be Outstanding AND Innovative AND creative AND influential. That seems hard to do :-(  If they made it an OR instead of an AND I may ask someone to nominate me for my Muffin Work. It does use 0-1 programming!

3) The past winners are, of course, very impressive. But there is one I want to point out to emphasize that the scope is broad: Amit Sahai won in 2022, and here is what the webpage says about it:

For a leading role in development of cryptographic Software Obfuscation and its applications, starting from initial conception of "Indistinguishability Obfuscation" and culminating in new constructions based upon well-founded cryptographic assumptions. These breakthroughs highlight how computational complexity can enable secrecy while computing in insecure environments.

4) Comb Opt and Discrete Opt seem to be Operations Research. So this inspires the following question:

What are the similarities and differences between Operations Research and Research on Algorithms? 

I tend to think of Operations Research  as being more tied to the real world- but is that true?

5) Not enough 2-letter combinations  for what you want to say: I had to use the term Operations Research instead of the abbreviation OR since I was using OR for or. Oh well. 

Juris Hartmanis passed away on July 29 at the age of 94

 Juris Hartmanis, one of the founders of Complexity Theory, passed away on July 29 at the age of 94.  The Gödel's Last Letter blog has an obit posted here.  Scott Aaronson has some words and a guest post by Ryan Williams here. When other bloggers post obits I will update this paragraph to point to them. 

Here is one non-blog obit: here.  For an interview with Juris see here.

Hartmanis and Stearns shared the 1993  Turing award for the paper On the Computational Complexity of Algorithms (see here for the paper and see here for his Turing Award Talk). In that paper they define DTIME(T(n)) for Turing Machines. They also proved some theorems about how changes to the model (1-tape, 2-tape, 1-dim, 2-dim others) change the notion of DTIME(T(n))- so they were well aware that the definition was not robust. They also have some theorems about computable numbers. 

We are used to the notion that you can measure how long a computation takes my counting the number of steps on a Turing Machine. Before the Hartmanis-Stearns paper this was not how people thought of things. Knuth (I suspect independently) was looking at what we might now call concrete complexity- how many operations does an algorithm need. Hartmanis-Stearns were beginning what is now called Complexity Theory. 

 Recall that later, the Cook-Levin Theorem used Turing Machines. 

If Hartmanis-Stearns did not start the process of putting complexity on a rigorous mathematical basis how might complexity theory have evolved? It is possible we would not have the Cook-Levin Theorem or the notion of NP. It is possible that we would ASSUME that SAT is hard and use that to get other problems hard, and also do reverse reductions as well to get some problems equivalent to SAT. Indeed- this IS what we do in other parts of theory with assuming the following problems are hard (for various definitions of hard): 3SUM, APSP, Unique Games. And in Crypto Factoring, DL, SVP, and other problems. 

Hartmanis did other things as well. I list some of them that are of interest to me - other people will likely list other things of interest to them. 

0) He had 21 PhD Students, some of them quite prominent. The list of them is here.

1) The Berman-Hartmanis Conjecture: All NP-Complete sets are poly isomorphic. Seems true for all natural NP-complete sets. Still open. This conjecture inspired a lot of work on sparse sets including that if a sparse set S is btt-hard for NP, then P=NP (proven by Ogiwara-Watanabe)

2) The Boolean Hierarchy: we all know what NP is. What about sets that are the difference of two NP sets? What about sets of the form A - (B-C) where A,B,C are all in NP? These form a hierarchy. We of course do not know if the hierarchy is proper, but if it collapse then PH collapses.

3) He helped introduce time-bounded Kolmogorov complexity into complexity theory, see here.

4) He was Lance Fortnow's undergraduate advisor. 

5) There is more but I will stop here.

100 Best Number Theory books of all Time---except many are not on Number Theory

I was recently emailed this link:

100 Best Number Theory books of all Time

That sounds like a good list to have!  But then I looked at it. 

The issue IS NOT that the books on it are not good. I suspect they all are.

The issue IS that many of the books on the list are not on Number Theory.

DEFINITLY NOT:

A Mathematicians Apology by Hardy

The Universe Speaks in Numbers by Farmelo (looks like Physics)

Category theory in Context by Riehl

A First Course in Mathematical logic and set theory by O'Leary

Astronomical Algorithms by Meeus (Algorithms for Astronomy)

Pocket Book of Integrals and Math Formulas by Tallardia

Entropy and Diversity by Leinster

BORDERLINE:

Too many to name, so I will name categories (Not the type Riehl talks about)

Logic books. Here Number Theory  seems to mean Peano Arithmetic and they are looking at what you can and can't prove in it. 

Combinatorics books:  Indeed, sometimes it is hard to draw the line between Combinatorics and Number Theory, but I still would not put a book on Analytic Combinatorics on a list of top books in Number Theory. 

Discrete Math textbooks: Yes, most discrete math textbooks have some elementary number theory in them, but that does not make them number theory books.

Abstract Algebra, Discrete Harmonic Analysis, other hard math books: These have theorems in Number Theory as an Application.  But they are not books on number theory. 

WHAT OF ALL THIS? 

Lists like this often have several problems

1) The actual object of study is not well defined.

2) The criteria for being good is not well defined.

3) The list is just one person's opinion. If I think the best math-novelty song of all time is William-Rowan-Hamilton (see  here) and the worse one is the Bolzano--Weierstrass rap (see here) that's just my opinion. Even if I was the leading authority on Math Novelty songs and had the largest collection in the world, its still just my opinion. (Another contender for worst math song of all time is here.)

4) Who is the audience for such lists? For the Number Theory Books is the audience ugrad math majors? grad math majors? Number Theorists? This needs to be well defined.

5) The list may tell more about the people making the list then the intrinsic qualify of the objects. This is more common in, say, the ranking of presidents. My favorite is Jimmy Carter since he is the only one with the guts to be sworn in by his nickname Jimmy, unlike  Bill Clinton (sworn in as William Jefferson Clinton- a name only used by his mother when she was mad at him) or Joe Biden (sworn in as Joseph Robinette Biden which I doubt even his mother ever used). My opinion may seem silly, but it reflects my bias towards nicknames, just as the people who rank presidents use their bias. 

What is known about that sequence

 In my last post I wrote:

---------------------------

Consider the recurrence

a_1=1

for all n\ge 2, a_n = a_{n-1} + a_{n/2}.

For which M does this recurrence have infinitely many n such that a_n \equiv  0 mod M?

I have written an open problems column on this for SIGACT News which also says

what is known (or at least what I know is known).  It will appear in the next issue.

I will post that open problems column here on my next post.

Until then  I would like you to work on it, untainted by what I know. 

----------------------------------------

I will now say what is known and point to the open problems column, co-authored with Emily Kaplitz and Erik Metz. 

If  M=2 or M=3 or M=5 or M=7 then there are infinitely many n such that a_n \equiv 0 mod M

If M\equiv 0 mod 4 then there are no n such that a_n \equiv 0 mod M

Empirical evidence suggests that

If M \not\equiv 0 mod 4 then there are infinitely many n such that a_n\equiv 0 mod M

That is our conjecture. Any progress would be good- for example proving it for M=9. M=11 might be easier since 11 is prime. 

The article that I submitted is HERE

An open question about a sequence mod M.

In this post n/2 means floor{n/2}

Consider the recurrence

a_1=1

for all n\ge 2, a_n = a_{n-1} + a_{n/2}.

For which M does this recurrence have infinitely many n such that a_n \equiv  0 mod M?

I have written an open problems column on this for SIGACT News which also says

what is known (or at least what I know is known).  It will appear in the next issue.

I will post that open problems column here on my next post.

Until then  I would like you to work on it, untainted by what I know. 

ADDED LATER: I have now posted the sequel which includes a pointer to the open problems column. To save you time, I post it here as well.

Review of The Engines of Cognition: Essays From the LessWrong Forum/Meta question about posts

 A while back I reviewed A Map that Reflects the Territory which is a collection of essays posted on the lesswrong forum. My review is here. I posted it to both this blog and to the lesswrong forum. In both cases I posted a link to it. My post to lesswrong is here

On the lesswrong post many of the comments, plus some private emails, told me NO BILL- don't post a link, post it directly as text. It was not clear how to do that, but I got it done with help.

On complexity blog nobody commented that this was a problem. Then again, nobody commented at all, so its not clear what to make of that. 

So

Meta Question: Is posting a link worse than posting direct text? Note that the book review was 12 pages long and looked great in LaTeX. 

Meta Question: Why did lesswrong care about the format but complexityblog did not (Probably answer: omplexityblog readers did not care at all, whereas Lesswrong cared about what I though about Lesswrong)

Another Question, not Meta. One of the comments was (I paraphrase)

When I open a pdf file I expected to see something in the style of an academic paper. This is written in very much chatty, free-flowing blog post style with jokes like calling neologisms ``newords'', so the whole think felt more off-kilter than was intended. The style of writing would prob work better as an HTML blog post (which could then be posted directly as a Lesswrong post here instead of hosted elsewhere and linked.)

I think its interesting that the format of an article telegraphs (in this case incorrectly) what type of article it will be. Is this a common problem?  I have had the experience of reading a real academic paper and being surprised that some joke or cultural-reference is in it, though I do not object to this. 

Another comment and question

I was surprised the post only had 11 karma when I saw it (William had send me an advance copy and I'd really liked reading it) but when I saw that it was a link post, I understood why.

I find this hilarious- they have some way the posts are rated!  For one, Lance told me very early on to never worry about comments, and I don't. Second, it reminds me of the Black Mirror episode Nosedive.

ANYWAY, I have reviewed another collection of essays for less wrong, this one called The Engines of Cognition. I am posting it here as a link: here  and I will post it on lesswrong as full text (with help) in a few days. 

I am posting it so I can get comments before I submit it to the SIGACT News book review column. But this is odd since I think this blog has more readers than SIGACT news has subscribers, so perhaps THIS is its real debut, not that. And of course the lesswrong forum is a place where more will read it since its about them. 

So- I appreciate comments to make it a better review!

The Highland Park Shooting

This week I should be celebrating Mark Braverman's Abacus Medal and the Fields Medalists. Instead my mind has been focused 25 miles north of Chicago.

Mass shootings in the United States have become far too commonplace, but the shooting at a fourth of July parade in Highland Park, Illinois hit home. Literally Highland Park was home for me, from 2003-2012. We've been in downtown Highland Park hundreds of times. We've attended their fourth of July parade in the past. My daughter participated in it as part of the high school marching band. 

We were members of North Shore Congregation Israel. My wife, who had a party planning business back then, worked closely with NSCI events coordinator Jacki Sundhein, tragically killed in the attack.

We lived close to Bob's Deli and Pantry and we'd often walk over there for sandwiches or snacks, sometimes served by Bob Crimo himself. The alleged shooter, Bobby Crimo, was his son.

We spent the fourth with friends who came down from Glencoe, the town just south of Highland Park. We spent much of the day just searching for updates on our phones.

I wish we could find ways to reduce the shootings in Highland Park and those like it, the violence that plagues Chicago and other major cities and the highly polarized world we live in which both hampers real gun reforms and creates online groups that help enable these awful events. But right now I just mourn for the lives lost in the town that was my home, a town that will never fully recover from this tragedy.

A Gadget for 3-Colorings

Following up on Bill's post earlier this week on counting the number of 3-colorings, Steven Noble emailed us with some updated information. The first proof that counting 3-colorings is #P-complete is in a 1986 paper by Nati Linial. That proof uses a Turing reduction using polynomials based on posets.

Steven points to a 1994 thesis of James Annan under the direction of Dominic Welsh at Oxford that gives the gadget construction that I so tried and failed to do in Bill's post.

Think of color 0 as false and color 1 as true and use this gadget in place of the OR-gadgets in the regular NP-complete proof of 3-coloring. I checked all eight values of a, b and c and the gadget works as promised.

James Annan later became a climate scientist and co-founded Blue Skies Research in the UK. 

Steven also noted that counting 2-colorings is easy, because for each connected component, there are either 0 or 2 colorings.

Update 4/22/24: We later learned of another gadget construction by Graham Farr in a 1994 Acta Informatica paper.