This document discusses parallel binary search. It divides a problem space from minimum to maximum into multiple sub-ranges that can be searched in parallel. The sub-ranges are (min, h), (h, h2), and (h2, max) where h and h2 are splitting points. Each processor then performs binary search on one sub-range independently. This allows the queries to be handled more quickly by searching multiple parts of the problem space in parallel rather than sequentially.
This document discusses parallel binary search. It divides a problem space from minimum to maximum into multiple sub-ranges that can be searched in parallel. The sub-ranges are (min, h), (h, h2), and (h2, max) where h and h2 are splitting points. Each processor then performs binary search on one sub-range independently. This allows the queries to be handled more quickly by searching multiple parts of the problem space in parallel rather than sequentially.
This document discusses algorithms and geometry. It covers the Postech Computer Algorithm Team and convex hulls. The Postech team works on algorithms and geometry problems. Convex hull concepts are explained, including finding the farthest point and defining a convex polygon. Formulas for computing areas and determining if a point is inside a convex shape are provided.
The document discusses the Fast Fourier Transform (FFT) algorithm and its use for multiplying two polynomials. It explains that FFT uses the Fourier transform to break down the polynomial multiplication problem into smaller subproblems by mapping the polynomials to the frequency domain using the Discrete Fourier Transform (DFT). It then describes the Cooley-Tukey algorithm, which is an efficient method for computing the DFT and allows FFT to multiply two polynomials in O(n log n) time rather than O(n^2) time for the standard multiplication approach.
The document discusses segment tree data structures for storing interval data. It describes persistent and dynamic segment trees, which allow maintaining interval data over time as the data is updated. The segment trees support operations like querying the total value in an interval or updating the value of an interval efficiently in logarithmic time.
The document describes a network flow algorithm. It involves modeling the network as a flow graph with nodes and edges having capacities and costs. The algorithm finds the maximum flow from a source to a sink in the graph while minimizing the total cost of the flow. It works by scaling all edge costs and finding augmenting paths with negative reduced costs to increment the flow over each path.
This document discusses concepts in number theory and combinatorics from the POSTECH Computer Algorithm Team. It covers topics like prime numbers, factorization, and counting principles. Example problems and their solutions are provided to illustrate concepts in number theory like finding the greatest common divisor and factoring numbers.