Completed
Machine Learning for Relevance and
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Machine Learning for Relevance and Serendipity
Automatically move to the next video in the Classroom when playback concludes
- 1 Machine Learning for Relevance and
- 2 The Discovery Problem
- 3 How we do it
- 4 People who use the term 'secret sauce' know about neither
- 5 Talk Outline
- 6 Our Topic Classifier
- 7 Interest Scores
- 8 Saved Session Data
- 9 You are all unique snowflakes
- 10 How do we know if we've done a good job?
- 11 Formula for a Machine Learning Problem
- 12 Collect Data • Save session information (described earlier)
- 13 Featurize Data
- 14 Training the model (numeric optimization)
- 15 Classify New Data
- 16 Original formulation weight vector is universal, features are user-specific
- 17 The Development Cycle
- 18 Data bugs (are the worst)
- 19 Presentation bias
- 20 Ranking vs Normal ML
- 21 Statistical Bleeding
- 22 Simpsons Paradox
- 23 Summary • We collect information about our users' preferences and dispreferences
- 24 Prismatic Backend Team
