Calendar

Oct
12
Fri
Colloquium- Yuriy Mileyko
Oct 12 @ 3:30 pm – 4:30 pm

Speaker: Yuriy Mileyko (UHM)
Title: Another look at recovering local homology from samples of stratified sets.

Abstract
Recovering homological features of spaces from samples has become one of
the central themes in topological data analysis, which has led to many
successful applications. Most of the results in this area focus on
global homological features, whose recovery predicates on imposing well
understood geometric conditions on the underlying space. Existing work
on recovering local homological information of a space from samples has
been much less abundant, since the required local geometric conditions on
the underlying space may vary from point to point and are not easily
integrated into a global condition, unless the space is a smooth
manifold. In this talk, we show that such global conditions for
recovering local homological information can be obtained for a fairly
large class of stratified sets.

Oct
15
Mon
Logic seminar: David Ross
Oct 15 @ 2:30 pm – 3:30 pm

Prof. Ross will speak about Nathanson’s “Generalized Egyptian Fractions.” Usual time and place (2:30, K314).

Abstract:
In a paper earlier this year Mel Nathanson generalized the notion of
“Egyptian Fraction” and extended some results of Sierpinski to sets of
these generalized Egyptian fractions. I’ll give short nonstandard proofs of
further generalizations of these results.

Oct
19
Fri
Colloquium: Ricardo Teixeira (U. Houston-Victoria) @ Keller 301
Oct 19 @ 3:30 pm – 4:30 pm

Speaker: Ricardo Teixeira (U. Houston-Victoria)
Title: Teaching Precalculus in Hawaii

Abstract: Most students enjoy learning about applications of math concepts. In
this talk, we will show how Precalculus concepts can motivate students
in the University of Hawai`i at Mānoa. Some interesting and recreational
use of certain concepts will be explored. At the end, we will cover
other ideas such as how to develop and maintain an effective culture of
assessment that may be used for future improvement and more.

Oct
29
Mon
Logic seminar: The number of maximally complex languages
Oct 29 @ 2:30 pm – 3:30 pm

Speaker: B. Kjos-Hanssen (joint work with Lei Liu)
Abstract:
Campeanu and Ho (2004) stated that it is “very difficult” to compute the number $m_n$ of maximally complex languages (in a finite automata sense) consisting of binary words of length $n$. We show that $m_n=O_{i,n}$, the number of functions from $[2^i]$ to $[2^{2^{n-i}}]$ whose range contains $[2^{2^{n-i}}-1]$, for the least $i$ for which $O_{i,n}>0$. Here, $[a]=${1,…,a}.

Nov
2
Fri
Colloquium John Marriott (Boeing) @ Keller 401
Nov 2 @ 3:30 pm – 4:30 pm

Speaker: John Marriott (Boeing)
Title: Data Science Curriculum for Industry

Abstract:

John Marriott earned his PhD from UH Math in 2013 and currently works
at Boeing as a data scientist. He combines mathematical modeling,
statistics, and programming to create data products on logistics,
labor estimates, and workplace safety. He will talk about his current
work, the transition from academia to industry, and suggestions for
curriculum to prepare students for work in this field.

Colloquium: Pamela Harris (Williams)
Nov 2 @ 3:30 pm – 4:30 pm
Nov
5
Mon
Jake Fennick: Probabilistic logic @ Keller 314
Nov 5 @ 2:30 pm – 3:30 pm

Title: Logic with Probability Quantifiers

Abstract: This talk is based on chapter XIV of Model-Theoretic Logics
(https://projecteuclid.org/euclid.pl/1235417263#toc). I will first give
a brief review of admissible sets and the infinitary logic which is
necessary for probability quantifiers. Then I will present the language
of probability quantifiers, as well as the proof theory, model theory,
and some examples which indicate the expressive power of the language.
Time permitting, my goal is to work towards the main completeness
theorem in section 2.3

Nov
16
Fri
Colloquium: Kameryn Williams (UHM) @ Keller 401
Nov 16 @ 3:30 pm – 4:00 pm

Speaker: Kameryn Williams (UHM)

Title: The universal algorithm, the $Sigma_1$-definable universal finite sequence, and set-theoretic potentialism

Abstract: As shown by Woodin, there is an algorithm which will computably enumerate any finite list you want, so long as you run it in the correct universe. More precisely, there is a Turing machine $p$, with the following properties: (1) Peano arithmetic proves that $p$ enumerates a finite sequence; (2) running $p$ in $mathbb N$ it enumerates the empty sequence; (3) for any finite sequence $s$ of natural numbers there is a model of arithmetic $M$ so that running $p$ in $M$ it enumerates $s$; (4) indeed, if $p$ enumerates $s$ running in $M$ and $t$ in $M$ is any finite sequence extending $s$, then there is an end-extension $N$ of $M$ so that running $p$ in $N$ it enumerates $t$. In this talk, I will discuss the universal algorithm, along with an analogue from set theory due to Hamkins, Welch, and myself, which we call the $Sigma_1$-definable universal finite sequence.

These results have applications to the philosophy of mathematics. Set-theoretic potentialism is the view that the universe of sets is never fully completed and rather we only have partial, ever widening access. This is similar to the Aristotelian view that there is no actual, completed infinite, but rather only the potential infinite. A potentialist system has a natural associated modal logic, where a statement is necessary at a world if it is true in all extensions. Using the $Sigma_1$-definable universal finite sequence we can calculate the modal validities of end-extensional set-theoretic potentialism. As I will discuss in this talk, the modal validities of this potentialist system are precisely the theory S4.