contextual /

Filename Size Date modified Message
642 B
A selection of my favorite software tools for octave.
The octave code contains a demo of the linUCB learning algorithm
that I used in my master's thesis. The implementation follows the
pseudo code proposed in "A Contextual-Bandit Approach to Personalized 
News Article Recommendation", Lihong Li, Wei Chu, Yahoo! Labs.

Copyright: Malte Ahl

    |-bm25.m          Computes Okapi Best Match 25 values of a matrix.
    |-softmax.m       Computes softmax values for a matrix.
    |-linUCB.m        The linUCB learning algorithm used in my master's thesis.
    |-demo_linUCB.m   A demo of the linUCB algorithm.