Information and partitions
Build from uncertainty to exact small-system partition search, then read the normalized MIP study with the right definitions.
What you will be able to do
Compute elementary information quantities and distinguish a normalized partition objective from the algorithm used to search it.
Open a lesson to begin.
The lessons
- 01Measure uncertainty with entropyTurn a binary probability into a reproducible information quantity.Ready to learn12 min
- 02Measure shared informationConnect marginal and joint entropy without confusing matching labels with dependence.Ready to learn12 min
- 03Enumerate graph bipartitionsRepresent each nonempty split once and inspect which connections cross it.Ready to learn12 min
- 04Read the normalized MIP objectiveSeparate predictive information loss, its normalizer, and the selected partition.Ready to learn12 min
- 05Compare exact and heuristic searchUse an exhaustive reference to evaluate how often faster searches recover the chosen optimum.Ready to learn12 min
Progress is kept in this browser. How learning works