Completed
More Optimal Approach
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Machine Learning for Optimal Matchmaking
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 What is matchmaking?
- 3 What is Machine Learning?
- 4 Machine Learning and Matchmaking
- 5 State of the Art Today: Skill
- 6 Real-time matchmaking
- 7 First Configure a Set of Rules
- 8 Example Configuration
- 9 Matchmaker Receives Requests
- 10 Matchmaker Request
- 11 Compares Requests
- 12 Creates a Match
- 13 Rules Apply Globally
- 14 Predictability How often the better team wins
- 15 Conventional Takeaway
- 16 More Optimal Approach
- 17 Defining Optimal
- 18 High-level Comparison
- 19 Unified objective function
- 20 As an Equation
- 21 Utility Function Use
- 22 True Match algorithm
- 23 True Match components
- 24 Population tracker
- 25 Population model
- 26 Metric Predictor
- 27 Simplest wait time formula • matchable(t) = requests that can match with t
- 28 Parameterized wait time formula
- 29 Optimizer: How does it make Rules?
- 30 Start with Current Rules
- 31 Find Optimal Transform Need to rewrite to search all curves
- 32 We found it!
- 33 True Match Rule Example Scale = 10, remember gap of 1 is OK
- 34 Mapped vs. Conventional
- 35 True Match Rules
- 36 What about Regions?
- 37 Conventional Region Approach
- 38 True Match Approach
- 39 FFA example
- 40 Skill gap variation
- 41 Wait time variation
- 42 True Match FFA Take-away
- 43 True Match Takeaway
- 44 Simple Improvements Matchmake on scaled Skill Percentiles
- 45 Thank you! Questions?
