Posts Tagged ‘Computer Science’

Bernstein on Preventive Security

Friday, November 16th, 2007

Dan Bernstein (also known as djb, the mastermind who wrote QMail, recently wrote a paper discussing methods of ensuring security. The approach is unique: his notion of security is not based on posthumous investigation such as minimizing privileged code and intrusion detection systems, but on a methodology transcending all phases of engineering an application that employs methods that disallow the engineer to write code that can be exploited. It is located at http://cr.yp.to/qmail/qmailsec-20071101.pdf.

European Summer School in Information Retrieval (ESSIR 2007)

Sunday, June 17th, 2007

This year, the European Summer School in Information Retrieval will be held in Glasgow, Scotland, United Kingdom, hosted by the University of Glasgow.

The seminars taught are to give participants a grounding in the core subjects of Information Retrieval, such as Architecture, Information Retrieval Models and Evaluation as well as various important applications such as Web Information Retrieval, Interactive Information Retrieval, XML Retrieval, Question Answering and Multimedia Retrieval.

As one can tell, the relevant industry is «organizing the world’s information», which is why Google is one of the sponsors.

An Algorithm for Compressing Space and Time

Friday, July 21st, 2006

An Algorithm for Compressing Space and Time by Tomas G. Rokicki

Making a slow program fast can lead to both joy and frustration. Frequently, the best you can do is a low-level trick to double or maybe quadruple the speed of a program; for instance, many readers may have implemented John Conway’s “Game of Life” using bit-level operations for a significant speedup. But sometimes a whole new approach, combining just a few ideas, yields amazing improvements. A simple algorithm called “HashLife,” invented by William Gosper (”Exploiting Regularities in Large Cellular Spaces,” Physica 10D, 1984), combines quadtrees and memoization to yield astronomical speedup to the Game of Life. In this article, I evolve the simplest Life implementation into this algorithm, explain how it works, and run some universes for trillions of generations as they grow to billions of cells.

Functional Programming: an introduction

Saturday, June 24th, 2006

This article seems to be an enlightening one about functional programming, emphasizing on what motivates the features of functional programming languages using Java examples (!).

Interpolation

Sunday, April 16th, 2006

Δεδομένων Π ζευγών (N, t(N)), να βρεθεί η τάξη (η δύναμη) της συνάρτησης που παράγει τις τιμές αυτές αγνοώντας σταθερούς παράγοντες και όρους μικρότερης τάξης.

Μπορείτε να υποθέσετε ότι το N είναι το μέγεθος ενός dataset και το t(N) ο χρόνος που απαιτεί ένας αλγόριθμος για να το λύσει. Ζητούμενο είναι η πολυπλοκότητα O του αλγορίθμου.

Μικρότερη τάξη:

  • Το x^m είναι μικρότερης τάξης από ότι το x^n.
  • Το logx είναι μικρότερης τάξης από ότι το x.
  • Κάθε σταθερός αριθμός είναι μικρότερης τάξης από το logx.
  • H ιδιότητα αυτή είναι μεταβατική: Αν το f(x) είναι μικρότερη τάξης από το g(x), τότε το h(x)*f(x) είναι μικρότερης τάξης από το h(x)*g(x), για παράδειγμα εφόσον το 1 είναι μικρότερης τάξης από το logx, τότε το x είναι μικρότερης τάξης από το x*logx.

References:

Διόρθωση: Αντί για “Αν το f(x) είναι μικρότερη τάξης από το h(x)” το σωστό είναι “Αν το f(x) είναι μικρότερη τάξης από το g(x)”.


^