Just Jim

James Harris Simons just passed away—he preferred Jim so we will use that here. I know that there will be here articles, posts, and special places to read about his wonderful life and work. So I thought I might add something that will be a bit different. I hope he would have liked this if he was still around.

Here is a quote of his that I like quite a bit:

I wasn’t the fastest guy in the world. I wouldn’t have done well in an Olympiad or a math contest. But I like to ponder. And pondering things, just sort of thinking about it and thinking about it, turns out to be a pretty good approach.

One can predict the course of a comet more easily than one can predict the course of Citigroup’s stock. The attractiveness, of course, is that you can make more money successfully predicting a stock than you can a comet.

IDA and More

I interacted with people who interacted much more closely with Jim that I ever did.

In 1964, Jim worked with the National Security Agency to break codes. Between 1964 and 1968, he was on the research staff of the Communications Research Division of the Institute for Defense Analysis (CRD of IDA). After being forced to leave IDA due to his public opposition to the Vietnam War, he worked on other problems. He founded Renaissance Technologies, a quantitative hedge fund. It eventually made huge returns so Jim was described as the “greatest investor on Wall Street”, and more specifically “the most successful hedge fund manager of all time”.

But I had no connection to Renaissance Technologies. But I did have a connection to him via IDA. Some of our friends in CS theory such as

Don Coppersmith
Donald Knuth
Nick Patterson
J. Barkley Rosser

also consulted at IDA in Princeton. When I was first at Princeton I was asked to join IDA—this was helped by Bob Sedgewick who had also had just joined Princeton. Bob had already been working as a consultant with IDA. Bob was terrific since while he was a theorist he definitely could write real programs.

This led me to get to meet Lee Neuwirth then the director. He was really a great leader: Neuwirth was the Deputy Director at IDA/CRD during the anti-war, anti-IDA demonstrations at Princeton, later serving as Director for eight years. He worked at IDA until his retirement in 1999. He is the recipient of the Exceptional Career Service Award from the National Security Agency. Nick Patterson later served also as the Director after he initially worked for the British code-breaking agency here.

Neuwirth wrote a book on his personal times at IDA. This is a bit challenging since the work done there was—well I cannot say much about it.

Open Problems

I hope this added some to the tidal wave of articles on Jim. I thought a bit about saying I knew that hedge fund group solved P=NP but thought I might get in trouble making that claim. Oh well perhaps they did?

COMPUTATION, LOGIC, AND INFORMATION

The organizers of a conference on this topic is on July 3-5, 2024 at Jagiellonian University, Krakow, Poland are:

Wojciech Szpankowski the Director of the Center for Science of Information at Purdue University

Marek Zaionc (UJ, Poland) the Professor and Head of the Foundations of Computer Science group at Jagiellonian University

See the poster of the conference that is on logic and related topics—it looks wonderful and I hope many can be there.

The Conference

Here is a link that gives all the information you need about it. The hope is that you will be able to attend this exciting event. Here is the lead in on the viewpoint of the conference:

Information, computation, and logic are defining concepts of the modern era. Shannon laid the foundation of information theory, demonstrating that problems of communication and compression can be precisely modeled, formulated, and analyzed. Turing formalized computation defined as the transformation of information by means of algorithms. Godel established modern foundation of logic, laying the foundation for modern computer science and science of information.

Open Problems

I could not list all the talks. But here is Avi Wigderson’s talk on errors in proofs. Just a sample of the interesting topics covered.

Frontiers in Complexity Theory

This summer DIMACS will hold an advanced workshop for graduate students in complexity theory. It is on July 29 till August 1. The organizers are: Lijie Chen, Roei Tell, Ryan Williams.

Their goal is to bring together up-and-coming complexity researchers and to introduce them to some recent exciting lines of research in the field.

Workshop attendance does not require specific prior knowledge, beyond interest in complexity theory and mathematical maturity. In particular, the workshop is also open to strong students working in other areas of theoretical computer science who are interested in learning more about recent complexity research.

The workshop will include keynote lectures by Avi Wigderson (IAS) and Ryan Williams (MIT). Also planned are:

Meta-complexity, taught by Rahul Ilango (MIT).
Error-correcting codes, taught by Swastik Kopparty (U Toronto).
Algebraic complexity, taught by Nutan Limaye (IT University of Copenhagen).
Derandomization, taught by Roei Tell (U Toronto).

A presentation of the new C3 locally testable code, by Lijie Chen is also planned.

Open Problems

The size of the event is limited — please visit here for instructions about applying.

1968 vs 2024

As we know, the last few days, have been tough times for certain groups of students at our universities. The groups have rioted, were then arrested by police, and then been handcuffed and led away to jail.

Mass arrests among students are quite special since such events are extremely rare. Examples of a such events happen in 1968—over 56 years ago. This was the so called Vietnam War riots that were famous when students demonstrated against the it.

2024

The riots today at some of our best places for education have been compared to those from previous times—such as the 1968 ones for example. See here and here.

1968 Again?

Is today, 2024, another 1968? Our key point is it is not:

While both times have had riots of students at university campuses we believe that today is fundamentally different.

How do 1968 and 2024 differ? We feel that understanding how they are really different, and how it yields insights should help us see how we can change the responses. This should help us do a better job handling the fundamental problems today. See here.

Differences

We are math oriented here—we are not experts on history nor on current events. But our viewpoint is that the structure of 1968 and 2024 cases reveals important insights. We hope you agree:

Case 1968: Here the parties that are against each other is simple. It’s the police and the protesters. So we note that we can call this the case when n=2. Here n measures the number of coherent groups involved.

Case 2024: Here the parties that are against each other is more complex. It’s the police against the two types of protesters: the pro Palestine and the pro Israel. So we note that we can call this the case when n=3.

When a parameter n varies from n=2 to n=3 a problem change in a basic manner. When the fundamental objects vary in number—especially from 2 to 3 all changes. Sometimes problems stop having simple solutions; or they stop having solutions at all. This is the case when its the two, and then the three

Two vs Three Body Problems

Do you think the view of 1968 being different from 2024 is related to the two vs three body problems? I think it explains to some degree why things seem to be much more complex today. If you agree does it yield a helpful view? Or is it completely silly? I think that it helps me understand why things today are more tricky? Do some of you agree?

Open Problems

I must end with that Kathryn Farley–my dear wife—and I have spent whole evenings watching the terrible events as they unfolded on TV. We are so happy that for all the tough times as far as we know no one has been hurt in any serious way. The biggest issue seems to be the cancelling of classes and graduation. We hope that it gets no worse in the future.

We hope that times improve and become quieter in the future. All stay safe.

Science Section

Alice says “How long is forever?” “Sometimes just one second.” Replies the White Rabbit

Tuesday is a special day for the New York Times—since it always contains the Science Section. This section is current exciting articles on science. We wonder if Alice would find it fun to read—I always love to read it. I hope you do too. David Corcoran was the editor of Science Times and you can read his book on some of the top articles from it here.

Science—A Problem

A sample of an article that you can find in the Science Section is here:

Generative A.I. Arrives in the Gene Editing World of CRISPR.
By Cade Metz

The trouble with all of the Science Sections is that they usually cover lots of stuff, but rarely if ever do they cover anything mathematical or anything complexity theory. This has lead us to make up our own Science Section that violates their rule. It will have lots of math and computer science and even complexity theory. I hope you will enjoy it.

Science Section

Here is a fake article that could be in our version of the science section. I hope you like it.

Cicadas Are Emerging Now. How Do They Know When to Come Out? see
here.

Scientists are making computer models to better understand how the mysterious insects emerge collectively after more than a decade underground. They are the cicadas and act on cyclic basis: The Great Southern Brood, which emerges across the South and the Midwest every 13 years, has been seen at sites scattered from North Carolina to Georgia. The Northern Illinois Brood, which appears every 17 years in the Midwest.

This has led many to study the cyclic structure of boolean functions. We highlight the interesting paper paper Testing Odd-Cycle-Freeness in Boolean Functions by Arnab Bhattacharyya, Elena Grigorescu, Prasad Raghavendra, Asaf Shapira. They show see …

Open Problems

Do you like our pretend article for the science section?

An Open Problem

Richard Feynman and Gian-Carlo Rota worked on different parts of science during their separate careers. Feynman of course was one of the most important scientists of the 20th century—see here.

His work in theoretical physics radically reshaped our understanding of the universe we live in at the most fundamental subatomic levels.

But Feynman’s brilliance was not solely due to his natural cognitive abilities. He relied on a method: a simple technique for seeing the world through the lens of open-ended questions, which he called his “favorite problems.”

Rota worked in the subject of combinatorics; and he lifted it from a barely respectable obscurity to one of the most active areas of mathematics today—to quote Richard Stanley, one of his great students. Rota had many neat results including one of the great papers Ten Lessons I Wish I Had Been Taught is wonderful. It fits neatly with Feynman notion of problems.

The Problem

Our problem today is not one that Feynman nor Rota would have most likely have worked on. Not physics and not combinatorics. But I do think it is possibly one that is still interesting. I hope you agree.

We often have to create a post so that a user can ask some particular question. The post should make it easy for them to get the answer that they are interested in without any difficulty. This is not always easy for us to make it easy.

One simple issue is this: Given a post we wish to make it easy for you to be able to ask for some particular information. The problem is that the actual post may let you get into some type of nasty state. The state may be hard to get into a place where you feel you are in control. The post may be in some strange state—one that does not respond to you questions in a way that you understand. This can be quite upsetting and led to a tough experience for your audience. This is something that we wish to avoid.

More Details

Imagine you are using our post. You interact via some application programming interface (API). You input information to the API by typing some keys or clicking on some icons. Each such input causes the API to change and enter some potentially new state. The difficulty is that this state may be a wrong one. You would like some input to force the post to reset. That is the API should change to a “reset” or some other nice state. But nothing causes this to happen. Nothing.

How can we fix the API to make that happen when we wish? This is the problem that we would like to solve.

Open Problems

This problem does have a simple solution. But the solution is not that pleasant. We could agree that some input—if you type XXX—will always reset the API. But we often do not wish to reserve a fixed sequence of inputs to always cause a reset. What if we wish to be able to input XXX in some situation? This is the problem. How do we have a way out of trouble and yet do not make the inputs restricted? The answer is not that natural we believe. What would you do?

One idea is have a special sequence like XXX that makes the reset. It does mean that XXX cannot be used in any other situation. But we could agree that XXX is replaced with a special secret sequence, that is extremely unlikely to be used except in the reset case. Perhaps we could replace XXX with the Fibonacci sequence 0112358132134 or some other special unusual sequence. Does this make some sense. It would have to be known by users and they would have to operate like this:

If the API is in some nasty case, they would say to themselves “I better type the special sequence.” Oh I recall it is 0112358132134 and the API will always reset.

And the reset will happen. Neat?

Women in Math Research

Peter Gerdes is a mathematician working in computability theory a.k.a. recursion theory, a branch of mathematical logic studying what computers (aka Turing machines) could in principle compute. Or more accurately when does being given access to the solution of one kind of problem (aka an oracle) allow the computer to solve some other problem. Currently focused on research about the alpha-REA degrees.

He is also the maintainer of the rec-thy latex package designed to give a common set of basic commands for the working mathematician in computability theory.

His History

Peter received his Ph.D. from the Group in Logic and Methodology of Science at the University of California at Berkeley under Prof. Leo Harrington. The Group in Logic is a cross-departmental program between the Math and Philosophy departments with some membership from the Computer Science department. The Group in Logic ensures that it’s graduates have a background in both the philosophical and mathematical aspects of logic but his advanced training and thesis were the same as if he had been in the mathematics department.

See here. for more details on his work.

His Women

He has said about articles on women in math:

That’s some very nice mathematics (both what’s covered here and what’s in the rest of the article). However, I worry that the primary effect of highlighting it in an issue devoted to women in mathematics is to create a perception that female mathematicians aren’t doing interesting enough work to hold their own so need their own special issue.

If you’re concerned about representation and encouraging fledgling female mathematicians isn’t the better move and simply increase the number of women covered in normal content?

Peter had a good point about giving proper credit to women. Here are two that he supports quite strongly:

His wife Sharon Berry is a tenured philosophy professor at Ashoka University who recently published a book.

The headline news is that Berry advocates a version of set theory that, rather than taking set theory to be the study of a single hierarchy of sets which stops at some particular point we should instead interpret set theorists as making claims about what hierarchies are possible. If you want to learn more about her work visit

here.

Karen Lange is an associate Professor of Department of Mathematics at Wellesley College. Her interests are in computability theory, an area of logic that explores the algorithmic content encoded in mathematical problems. She studies the computational complexity of problems of all sorts.

Open Problems

Do you agree with Peter’s point?

Guggenheim 2024

Edward Hirsch is the President of the John Simon Guggenheim Memorial Foundation and has lead it since 2003. He just announced the appointment of 188 Guggenheim Fellowships to a distinguished and diverse group of culture-creators working across 52 disciplines. Chosen from a pool of almost 3,000 applicants, the Class of 2024 Guggenheim Fellows was tapped on the basis of prior career achievement and exceptional promise. As established in 1925 by founder Senator Simon Guggenheim, each fellow receives a monetary stipend to pursue independent work at the highest level under “the freest possible conditions.

Edward Hirsch is a poet and shows that the depth and breadth of Guggenheim’s is large, See here

The range of the 188 new appointments is from 84 different academic institutions, 41 states and the District of Columbia, and four Canadian provinces are represented in this year’s class of Fellows, who range in age from 28 to 89. Many Fellows’ projects engage with issues like disability activism, climate change, identity, incarceration, AI, and democracy.

Computer Science

Gene Tsudik is the 2024 representative for computer science. He is a Distinguished Professor of Computer Science at the University of California, Irvine (UCI). He obtained his PhD in Computer Science from USC in 1991. Before coming to UCI in 2000, he was at IBM Zurich Research Laboratory (1991-1996) and USC/ISI (1996-2000). His research interests include many topics in security, privacy and applied cryptography.

Tsudik is the only computer scientist to be awarded a Guggenheim Fellowship this year, and he intends to use his fellowship funding to bootstrap a new line of research on building IoT devices resilient against devastating large-scale malware infestations that have become all too common in recent years. See his https://arxiv.org/pdf/2309.03574.pdf} with Sashidhar Jakkamsetti and Youngil Kim.

He is also the author of the first crypto-poem published as a refereed paper. Perhaps this is why Edward Hirsch helped select him? Perhaps not? Tsudik also suffers from two incurable academic diseases: “Research ADHD” and “Munchausen-without-Proxy.” Congrats to him as a winner of a Guggenheim.

Open Problems

I enjoyed some of his different points about computing. See this:

There are two equivalence classes of idiots:
Those who yearn for the past that never was.
AND
Those who dream of the future that never will be.

Congrats again.

2023 Turing Award

Congrats to Avi Wigderson for his winning the 2023 Turing Award.

It places Avi in with some of the great theorists of all time. See for other winners:
Knuth, Rabin, Scott, Cook, Karp, Hopcroft, Tarjan, Hartmanis, Blum, Yao, Rivest, Shamir, Adleman, Valiant, Micali, Goldwasser, Aho, Ullman.

Some Comments

Avi was a student of mine back at Princeton in 1983. I do recall one interesting thing. Back in those days the P=NP? problem was relatively new, having been created by Cook in 1971. We had no idea how hard its resolution was going to be. I recall Avi saying to me: “I plan on working on P=NP for a while—a few weeks.” Not too long, but as you know it is still unresolved.

Avi did prove tons of great results on things near to P=NP. Especially wonderful has been his contributions to randomness and related issues. He won the Turing Award for his work on understanding how randomness can shape and improve computer algorithms.

Open Problems

Avi says:

The [Turing] committee fooled me into believing that we were going to have some conversation about collaborating, he says. When I zoomed in, the whole committee was there and they told me. I was excited, surprised and happy.

Very clear he is a wonderful person, who is quite modest.

The Coming Solar Eclipse

Last Tuesday, April 2nd, the New York Times had its weekly Science section.

Its focus was all about the eclipse that happens on Monday, April 8th. When it will exactly happen? How best to watch it? How to enjoy even if you are blind—can you still enjoy it? And more. One how-to article they’ve had up online for longer is this.

The Eclipse

The main issue with the eclipse is, when is it safe to watch it? It is not safe unless one watches it with some important restrictions. Direct viewing of the sun without using any sort of protection can cause permanent eye damage. The intense light from the sun during an eclipse can cause damage to the retina of the eye, leading to permanent vision loss. It’s important to use specially designed solar viewing glasses or other indirect viewing methods to safely observe a solar eclipse.

I started reading the Science articles to see if there was a simple answer to the question:

Is it safe to watch it on TV?

This seemed like a simple direct question. Perhaps I am too TV based—I spend a major fraction of my day watching the news and other programs every day on TV. So why not watch the solar eclipse this way?

I discovered quickly that the Science articles had lots of comments on how to safely watch the eclipse. But none said anything about: is it TV safe? Doing a Google search showed that there were two answers:

1. In fact, the absolute safest way to watch it is on TV.

2. Yes, it is not safe to look at a solar eclipse, even on TV, without proper eye protection.

However, there is often confusion about how to look at it safely during an eclipse, when the moon partially or completely covers the sun and it gets dark outside. Viewing the solar eclipse safely is paramount. In fact, the absolute safest way to watch it is on TV. Having said this, the TV camera shooting the sun CAN get permanent damage if not protected in the same way as our eyes. But nothing that will appear on a TV screen is remotely close to be able to damage your eyes.

Up in Buffalo

Ken and visiting friends and family will be in the path of totality. The paths of clouds, however, are too uncertain to tell what they will see. Over to Ken:

I bought welder’s glass at the time of the 1994 annular eclipse in Buffalo—which was clouded out. The key is that if you hold it up to a bright bare light bulb, you should still NOT see the bulb at all. Same with eclipse glasses. The glass worked well for the 2017 eclipse which was partial in Buffalo.

Debbie and I attended a lecture last Thursday at Nichols School where she teaches. The speaker, Jeffrey Linsky, made one point to add to the above advice: once the eclipse goes total, take the glasses off. It is safe to look directly at the sun’s corona behind the moon. Linsky also emphasized that the human eyes have better resolution than any camera for viewing and appreciating features such as prominences, wisps of the corona, and the surrounding dark sky.

Exactly at the instant totality ends, whip the glasses back on before attempting to view Baily’s beads or the diamond ring effect.

Watching the 2017 eclipse on TV was indeed deadly—for a P != NP claim.

Open Problems

So I believe that TV is okay, but not all seem to agree. So be careful with this eclipse, please.