CS7646 FINAL EXAM GRADED A+
1 of 52
Definition
Minimizer finds coefficients C0, C1, ...etc...
f(x) = mx + b
f(x) = C0*X + C1
-
Give this one a try later!
, KNN where K varies Optimize a Portfolio
Kernel Regression Parameterized model
Don't know?
2 of 52
Definition
-training is generall a 60/40% split but in cross
validation we slice data up into difference
chunks and train on different portions of 80%
Give this one a try later!
Cross validation External validation
Train Regression
Don't know?
3 of 52
Definition
Success depends on exploration of as much
of state and action space as possible
,-we do this by flipping coin twice
1) random action or pick argmax
2)if random action which random action
- helped via random action rate RAR
- at the beginning a high RAR will force us to
explore the states
Give this one a try later!
Q-Learning Gamma and Alpha Q Learning Random Action
Q-Learning Pros and Cons Q-Learning Procedure
Don't know?
4 of 52
Definition
Regression - try to make numerical prediction
Classification - classifying into one or several
types
Give this one a try later!
Regression LinReg vs KNN vs Decision Tree
vs Classification
Performance
, Correlation vs RMSE Options
Don't know?
5 of 52
Definition
finds parameters for a model. Take
data to get parameters and then throw
away data
Problems:
-noisy and uncertain - value to be found - but it
has to be accumulated over many trading
opportunites
-challenging to estimate confidence
-holding time/allocation is uncertain
-RL policy learning is better
Give this one a try later!
Linear regression
(parametric learning) Q Learning Random Action
Q-Learning K Nearest Neighbor (KNN /
Gamma and Alpha
instance based)
Don't know?
6 of 52
1 of 52
Definition
Minimizer finds coefficients C0, C1, ...etc...
f(x) = mx + b
f(x) = C0*X + C1
-
Give this one a try later!
, KNN where K varies Optimize a Portfolio
Kernel Regression Parameterized model
Don't know?
2 of 52
Definition
-training is generall a 60/40% split but in cross
validation we slice data up into difference
chunks and train on different portions of 80%
Give this one a try later!
Cross validation External validation
Train Regression
Don't know?
3 of 52
Definition
Success depends on exploration of as much
of state and action space as possible
,-we do this by flipping coin twice
1) random action or pick argmax
2)if random action which random action
- helped via random action rate RAR
- at the beginning a high RAR will force us to
explore the states
Give this one a try later!
Q-Learning Gamma and Alpha Q Learning Random Action
Q-Learning Pros and Cons Q-Learning Procedure
Don't know?
4 of 52
Definition
Regression - try to make numerical prediction
Classification - classifying into one or several
types
Give this one a try later!
Regression LinReg vs KNN vs Decision Tree
vs Classification
Performance
, Correlation vs RMSE Options
Don't know?
5 of 52
Definition
finds parameters for a model. Take
data to get parameters and then throw
away data
Problems:
-noisy and uncertain - value to be found - but it
has to be accumulated over many trading
opportunites
-challenging to estimate confidence
-holding time/allocation is uncertain
-RL policy learning is better
Give this one a try later!
Linear regression
(parametric learning) Q Learning Random Action
Q-Learning K Nearest Neighbor (KNN /
Gamma and Alpha
instance based)
Don't know?
6 of 52
