lecture1

No always bad commifee decisions endless group meenngs

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: he Nme –  Phone ­A ­Friend Correct 65% of the Nme –  Ask the Audience Correct 91% of the Nme 8 Crowds Are Smarter… •  Jelly ­Beans ­in ­the ­Jar Experiment –  Professor Jack Treynor ran the experiment in his class –  with a jar that held 850 beans –  the group esNmate was 871 –  only one of the 56 people in the class made a beFer guess 9 Are Crowds Smarter? •  No always –  Bad commiFee decisions –  Endless group meeNngs •  In this course, we focus on mechanisms that intend to make crowds smarter. 10 Structure of the Course •  Introductory lectures (6 lectures) –  This one, game theory (2), and basics of proper scoring rules and predicNon markets (3) •  Research Papers –  PredicNon Markets –  Crowdsourcing and User ­Generated Content –  Peer PredicNon •  One or two guest lectures 11 Prerequisites •  Math background is important! At least a basic course in linear algebra (such as M 21b, AM 21b, or equivalent) •  A course on probabiliNes and staNsNcs (STAT 110 or equivalent) •  An algorithm course (CS 124, or equivalent) •  Familiarity with the concept of raNonality. An AI course or an economics/game theory course. CS 186 and advanced course in algorithms, microeconomics, game theory are helpful but not required. 12 Reading Materials •  For each class, we have provided some reading materials. We ask you to read them and submit your comments by midnight before class. •  There will be reading quesNons for each class. –  Your comments should include good faith answers to these quesNons. –  The quesNons are designed to facilitate in understanding or to enc...
View Full Document

Ask a homework question - tutors are online