BMGT 394H #1: Intro

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/26

flashcard set

Earn XP

Description and Tags

Covers Introduction PPT and Pandas-DataWrangling

27 Terms

1
New cards

deep learning

subset of machine learning composed of algorithms that permit software to train itself to perform tasks, like speech and image recognition, by exposing multilayered neural networks to vast amounts of data

2
New cards

machine learning

subset of AI that includes abstruse statistical techniques that enable machines to improve at tasks with experience

3
New cards

artificial intelligence

any technique that enables computers to mimic human intelligence, using logic, if-then rules, decision trees, and machine learning (including deep learning)

4
New cards

data set

list of records with values for some predefined fields

5
New cards

x causes y only if all 3 hold

  1. x is correlated with y

  2. x comes before y in time

  3. association isn’t explained by confounding variable


6
New cards

gold standard for establishing causality is a

randomized experiment

7
New cards

t/f: random assignment balances confounders across groups, so any difference must be caused by the treatment

true

8
New cards

why do you convert categorical variables to dummy variables?

categorical variables can’t go straight into a regression

9
New cards

inner join

keeps only keys found in both tables; default

10
New cards

left join

keeps every row in the left table

11
New cards

right join

keeps every row from the right

12
New cards

outer join

keeps all rows from both, provides NaN when no match

13
New cards

what can you do with NaN?

  1. drop them (dropna) —> loses data

  2. fill them (fillna) —> e.g. with mean, can introduce bias


14
New cards

binning/discretization

turns continuous variable into categories

15
New cards

inplace=True

changes original DataFrame; otherwise, Pandas returns a modified copy

16
New cards
17
New cards
18
New cards
19
New cards
20
New cards
21
New cards
22
New cards
23
New cards

issues with AI

  • adversarial attack

  • fake content

  • algorithmic bias

  • interpretability

  • algorithmic collusion

  • violation of privacy

  • environmental issues

  • copyright issues


24
New cards

designed system

a personal writes the rules

25
New cards

learned system

the computer finds the rules itself from the data → black box

26
New cards

pivot table

rearranges data into a spreadsheet-like summary

27
New cards

crosstab

special pivot table that counts freqiuencies