Reputation
6,103
Top tag
Next privilege 10,000 Rep.
Access moderator tools
Badges
9 30
Newest
 Nice Answer
Impact
~70k people reached

Apr
30
answered Trigonometric inequality in an obtuse triangle
Apr
30
comment How to solve this trigo integral ? (sine)
wolframalpha.com/input/…
Apr
29
awarded  Nice Answer
Apr
29
answered Prove that $n^2(n^2+1)(n^2-1)$ is a multiple of $5$ for any integer $n$.
Apr
28
revised Prove that $a^2+b^2+c^2\geq [2(a-b)^2(b-c)^2(a-c)^2]^{1/3}$
added 7 characters in body
Apr
28
revised Prove that $a^2+b^2+c^2\geq [2(a-b)^2(b-c)^2(a-c)^2]^{1/3}$
added 86 characters in body
Apr
28
revised Prove that $a^2+b^2+c^2\geq [2(a-b)^2(b-c)^2(a-c)^2]^{1/3}$
added 86 characters in body
Apr
28
answered Prove that $a^2+b^2+c^2\geq [2(a-b)^2(b-c)^2(a-c)^2]^{1/3}$
Apr
26
reviewed Approve If $A$ is positive definite then so is $A^k$
Apr
9
accepted Compute the expectation of steps making $n$ different balls the same
Apr
8
awarded  Nice Question
Apr
3
comment Compute the expectation of steps making $n$ different balls the same
Your comment are very helpful. Thanks a lot. I guess now I get it.
Apr
2
comment Compute the expectation of steps making $n$ different balls the same
ma.huji.ac.il/hart/papers/n-colors.pdf
Apr
1
comment Compute the expectation of steps making $n$ different balls the same
It seems not true for $n=4$, never mind.
Apr
1
comment Compute the expectation of steps making $n$ different balls the same
I conjecture that $E_{1^n}=E_{1^{n-1}}+E_{1^{n-1}2}$. It is not hard to solve $E_{1^{n-1}2}$. Naively, we first make the first $n-1$ balls the same color then the rest is $E_{1^{n-1}2}$.
Apr
1
revised A Fourier Analysis Question I am stuck at
added 86 characters in body
Apr
1
answered A Fourier Analysis Question I am stuck at
Apr
1
revised Compute the expectation of steps making $n$ different balls the same
added 4 characters in body
Apr
1
revised Compute the expectation of steps making $n$ different balls the same
deleted 1 character in body
Mar
31
comment About the definition of Day's convolution
coend is the 'integration' over the diagonal, so you need $c=c', d=d'$.