Please enable JavaScript.
Coggle requires JavaScript to display documents.
CSC311 Topics (Unsupervised Learning: no labeled examples – instead,…
CSC311 Topics
Unsupervised Learning: no labeled examples – instead, looking
for interesting patterns in the data
-
-
-
Decision Tree
Question: For entropy calculation, could it sum up to 2^n for each sublevel?
Also, does locally split on the most informative information criterion necessarily give the best global classification - based on just entropy alone?
-
-
-
random forest to help reduce correlation between data to "actually" reduce variance (sometimes it might be futile without introducing randomness to the choice to prevent overfitting)
-
-
-
-
-
-
-
-
-
Reinforcement Learning: : learning system receives a reward
signal, tries to learn to maximize the reward signal
-