אז בעזרת אפשר להוכיח ש:
א.
ב.
ג.
ד.
שאלה 2
נרצה לתת הגדרה חדשה עבור המחלקה NEXP, שתהיה שקולה לזו שהוצגה בהרצאה:
NEXP היא מחלקת השפות L{0,1}* עבורן קיימים מכונת טיורינג דטרמיניסטית M
ופולינום p, כך שלכל x{0,1}* מתקיים: xL אם ורק אם קיים w{0,1}*
באורך (1) כך ש- M(x,w)=1, וכן M רצה על (x,w) בזמן
(2) .
מהם החלקים החסרים בהגדרה זו?
א. (1). p(|x|) (2). p(|x|+|w|)
ב. (1). 2p(|x|) (2). p(|x|)
ג. (1). p(|x|) (2). 2p(|x|+|w|)
ד. (1). 2p(|x|) (2). p(|x|+|w|)
ה. (1). p(|x|) (2). 2p(|x|)
ו. (1). 2p(|x|) (2). 2p(|x|+|w|)
שאלה 3
טענה: קיים k>0 עבורו NP DTIME(nk).
את איזה משפט מהמשפטים הבאים אנו יכולים להוכיח?
א. הטענה נכונה.
ב. הטענה אינה נכונה.
ג. הטענה שקולה לטענה P=NP.
ד. הטענה שקולה לטענה NP=coNP.
Oded answered me via mail since I didn't have access to the forum.
Anyway if someone is interested these are his words:
1) Notice that the running time of the universal machine is
sqrt(SIZE OF ITS INPUT) * the running time of the M on x.
This means that given a machine that run in time n^2006,
the universal machine needs time n^{2006.5} because the
size of its input is some constant (the description of M)
plus the size of x (which is n).
2)Since w is of length exponential in x, allowing M to run in time
exponential in w, means that it can run in time *doubly exponential*
in x, which is believed to be much stronger than NEXP.
In the correct answer, the running time we allow is polynomial
in |x|+|w|, but since |w| can be exponential in |x|, this
can be exponential in |x|, as in the original definition.
(to prove this formally requires some more work, but I hope
the point is clear)
3)The statement there *implies* that P=NP but is not known to be
*equivalent*.
It can be shown to be FALSE because we know that for all k, DTIME
(n^2k)
is contained in NP but not in DTIME(n^k).