6,465 reputation
12064
bio website math.haifa.ac.il/~seva
location Israel
age 52
visits member for 4 years, 7 months
seen 1 hour ago
.

1d
awarded  Revival
1d
revised Minimal “sumset basis” in the discrete linear space $\mathbb F_2^n$
added 39 characters in body
1d
revised Minimal “sumset basis” in the discrete linear space $\mathbb F_2^n$
added 23 characters in body; edited tags
2d
answered Minimal “sumset basis” in the discrete linear space $\mathbb F_2^n$
May
7
awarded  Nice Answer
May
4
comment Vertex expansion of the Hamming graph
@Gordon Royle: seems you are right about the terminology. The graph in my answer is rather called the lattice graph, or the grid graph.
May
3
comment Vertex expansion of the Hamming graph
Still, check Bezrukov's survey. He addresses lots of variations. For $d=3$ there is no difference, BTW.
May
3
answered Vertex expansion of the Hamming graph
May
3
comment Positive roots of a polynomial
What makes you tho expect that it has exactly one positive root?
Apr
27
comment Is the set $ AA+A $ always at least as large as $ A+A $?
If $A=\{0,1\}$, then $AA=A$. So, one must assume something like $|A|\ge 3$.
Apr
27
revised Size of smallest multiple $\sum \epsilon_i p^i$ of $n$ with digits $\epsilon_i\in \{0,1\}$ for $p$ a prime
added 661 characters in body
Apr
26
answered Size of smallest multiple $\sum \epsilon_i p^i$ of $n$ with digits $\epsilon_i\in \{0,1\}$ for $p$ a prime
Apr
24
comment Size of smallest multiple $\sum \epsilon_i p^i$ of $n$ with digits $\epsilon_i\in \{0,1\}$ for $p$ a prime
In fact, it seems to me now that I was wrong. The congruence $p^N\equiv1\pmod n$ does not automatically imply $n\mid1+p+\dotsb+p^{N-1}$ as we do not assume $(n,p-1)=1$.
Apr
23
comment Size of smallest multiple $\sum \epsilon_i p^i$ of $n$ with digits $\epsilon_i\in \{0,1\}$ for $p$ a prime
Perhaps, a more appealing form of this question is this: for co-prime integer $p,n>1$, what is the smallest multiple of $n$ written to the base $p$ with the digits $0$ and $1$ only?
Apr
23
comment Size of smallest multiple $\sum \epsilon_i p^i$ of $n$ with digits $\epsilon_i\in \{0,1\}$ for $p$ a prime
The question would have a simple answer if you were allowing $-1$ as a coefficient. In this case, for $k\ge\log_2n$, at least two of the sums $\epsilon_0+\epsilon_1p+\dotsb+\epsilon_kp^k$ are congruent modulo $n$, and their difference is an algebraic sum of powers of $p$, divisible by $n$. Thus, you can take $\alpha=\log_2 p$ in this case.
Apr
20
comment Ordering subsets of the cyclic group to give distinct partial sums
@Pace Nielsen: This cannot happen as $P$ is a homogeneous polynomial; and so, no monomial contained in $P$ is dominated by another one. Does this answer your question?
Apr
17
awarded  Necromancer
Apr
17
awarded  additive-combinatorics
Apr
17
awarded  nt.number-theory
Apr
16
comment Ordering subsets of the cyclic group to give distinct partial sums
Hm-m-m... All the coefficients being even means that $P$ is the zero polynomial in ${\mathbb F}_2$, which is wrong?