Theo bài này ở tờ Science (đọc ở đây nếu bị chặn)
Two Saudi institutions are aggressively acquiring the affiliations of overseas scientists with an eye to gaining visibility in research journals.
At first glance, Robert Kirshner took the e-mail message for a scam. An astronomer at King Abdulaziz University (KAU) in Jeddah, Saudi Arabia, was offering him a contract for an adjunct professorship that would pay $72,000 a year. Kirshner, an astrophysicist at Harvard University, would be expected to supervise a research group at KAU and spend a week or two a year on KAU’s campus, but that requirement was flexible, the person making the offer wrote in the e-mail. What Kirshner would be required to do, however, was add King Abdulaziz University as a second affiliation to his name on the Institute for Scientific Information’s (ISI’s) list of highly cited researchers.
“I thought it was a joke,” says Kirshner, who forwarded the e-mail to his department chair, noting in jest that the money was a lot more attractive than the 2% annual raise professors typically get. Then he discovered that a highly cited colleague at another U.S. institution had accepted KAU’s offer, adding KAU as a second affiliation on ISIhighlycited.com.
Kirshner’s colleague is not alone. I have learned of more than 60 top-ranked researchers from different scientific disciplines—all on ISI’s highly cited list—who have recently signed a part-time employment arrangement with the university that is structured along the lines of what Kirshner was offered. Meanwhile, a bigger, more prominent Saudi institution—King Saud University in Riyadh—has climbed several hundred places in international rankings in the past 4 years largely through initiatives specifically targeted toward attaching KSU’s name to research publications, regardless of whether the work involved any meaningful collaboration with KSU researchers.
Một giáo sư nhận xét rằng
Neil Robertson, a professor emeritus of mathematics at Ohio State University in Columbus who has signed on, says he has no concerns about the offer. “It’s just capitalism,” he says. “They have the capital and they want to build something out of it.” Another KAU affiliate, astronomer Gerry Gilmore of the University of Cambridge in the United Kingdom, notes that “universities buy people’s reputations all the time. In principle, this is no different from Harvard hiring a prominent researcher.”
)
. Stanford vừa thông báo họ sẽ mở 
Loading ...
cũng là đa thức. Đa thức đối với kích cỡ của đầu vào của bài toán (ở đây sẽ là chiều của mô hình = số lượng tham số), và đa thức đối với 
. Epsilon ở đây là lỗi ước lượng cho phép. Ví dụ như 
vs. 
chẳng hạn.) Kỳ thực so với thống kê thì các kết quả về phức tạp mẫu bên computational learning theory còn khá đơn giản, vì họ chưa chú trọng đến độ đa thức. 
là một tốc độ kinh điển trong thống kê tham số hữu hạn chiều (khi đó, ta cũng có 
). Trong các vấn đề ước lượng khác thì tốc độ hội tụ có thể rất khác, 
hoặc rất chậm theo tốc độ logarithm, ![[log(1/n)]^{\delta}](http://s.wordpress.com/latex.php?latex=%5Blog%281%2Fn%29%5D%5E%7B%5Cdelta%7D%20&bg=ffffff&fg=000000&s=0 "[log(1/n)]^{\delta}")
, chẳng hạn. Lý thuyết minimax, nếu có, cho ta biết giá trị tốt nhất có thể của delta đạt được bởi một thuật toán tốt nhất có thể được. 