3 Evaluation criteria
Once we have specified a problem class, we need to say what makes an output or the an-
swer to a query good, given the training data. We specify evaluation criteria at two levels:
how an individual prediction is scored, and how the overall behavior of the prediction or
estimation system is scored.
The quality of predictions from a learned model is often expressed in terms of a loss
function. A loss function L(g, a) tells you how much you will be penalized for making a
guess g when the answer is actually a. There are many possible loss functions. Here are
some frequently used examples:
• 0-1 Loss applies to predictions drawn from finite domains.
If the actual values are
drawn from a contin-
uous distribution, the
probability they would
ever be equal to some
predicted g is 0 (except
for some weird cases).
If the actual values are
drawn from a contin-
uous distribution, the
probability they would
ever be equal to some
predicted g is 0 (except
for some weird cases).
L
(g
, a) =
0 if g = a
1 otherwise
• Squared loss
L
(g
, a) = (g − a)
2
• Linear loss
L
(g
, a) = |g − a|
• Asymmetric loss Consider a situation in which you are trying to predict whether
someone is having a heart attack. It might be much worse to predict “no” when the
answer is really “yes”, than the other way around.
L
(g
, a) =
1
if g = 1 and a = 0
10 if g = 0 and a = 1
0
otherwise
Any given prediction rule will usually be evaluated based on multiple predictions and
the loss of each one. At this level, we might be interested in:
• Minimizing expected loss over all the predictions (also known as risk)
• Minimizing maximum loss: the loss of the worst prediction
• Minimizing or bounding regret: how much worse this predictor performs than the
best one drawn from some class
• Characterizing asymptotic behavior: how well the predictor will perform in the limit
of infinite training data
• Finding algorithms that are probably approximately correct: they probably generate
a hypothesis that is right most of the time.
There is a theory of rational agency that argues that you should always select the action
that minimizes the expected loss. This strategy will, for example, make you the most money
in the long run, in a gambling setting. Expected loss is also sometimes called risk in the
Of course, there are
other models for ac-
tion selection and it’s
clear that people do not
always (or maybe even
often) select actions that
follow this rule.
Of course, there are
other models for ac-
tion selection and it’s
clear that people do not
always (or maybe even
often) select actions that
follow this rule.
machine-learning literature, but that term means other things in economics or other parts
of decision theory, so be careful...it’s risky to use it. We will, most of the time, concentrate
on this criterion.
Last Updated: 08/04/21 21:06:54
MIT 6.036
Fall 2021
9
Dostları ilə paylaş: |