some applications in human behavior modeling (and open questions)
TRANSCRIPT
![Page 1: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/1.jpg)
Some applications in human behavior modeling(and open questions)
Jerry Zhu
University of Wisconsin-Madison
Simons Institute Workshop on Spectral Algorithms:From Theory to Practice
Oct. 2014
1 / 30
![Page 2: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/2.jpg)
Outline
Human Memory Search
Machine Teaching
2 / 30
![Page 3: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/3.jpg)
Verbal fluency
Say as many animals as you can without repeating in one minute.
3 / 30
![Page 4: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/4.jpg)
Semantic “runs”1. cow, horse, chicken, pig, elephant, lion, tiger, porcupine,
gopher, rat, mouse, duck, goose, horse, bird, pelican, alligator,crocodile, iguana, goose
2. elephant, tiger, dog, cow, horse, sheep, cat, lynx, elk, moose,antelope, deer, tiger, wolverine, bobcat, mink, rabbit, wolf,coyote, fox, cow, zebra
3. cat, dog, horse, chicken, duck, cow, pig, gorilla, giraffe, tiger,lion, ostrich, elephant, squirrel, gopher, rat, mouse, gerbil,hamster, duck, goose
4. cat, dog, sheep, goat, elephant, tiger, dog, deer, lynx, wolf,mountain goat, bear, giraffe, moose, elk, hyena, aardvark,platypus, lion, skunk, wolverine, raccoon
5. dog, cat, leopard, elephant, monkey, sea lion, tiger, leopard,bird, squirrel, deer, antelope, snake, beaver, robin, panda,vulture
6. deer, muskrat, bear, fish, raccoon, zebra, elephant, giraffe,cat, dog, mouse, rat, bird, snake, lizard, lamb, hippopotamus,elephant, skunk, lion, tiger
7. dog, cat, ferret, fish, cow, horse, pig, sheep, elephant, tiger,lion, bear, giraffe, bird, groundhog, ox
8. antelope, baboon, cat, dog, elephant, frog, giraffe,hippopotamus, jaguar, dog, cat, horse, pig, chicken, bird,hippopotamus, cow
9. dog, cat, cow, sheep, lamb, rooster, chicken, tiger, elephant,monkey, orangutan, lemur, crocodile, giraffe, owl, bird, lion,elephant, hippopotamus, rhinoceros, penguin, seal, walrus,whale, dolphin
10. dog, cat, bear, deer, puma, opossum, moose, bear, wolf,coyote, lion, tiger, elephant, giraffe, hyena, cougar, mountainlion, antelope, elk
4 / 30
![Page 5: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/5.jpg)
Memory search
−1.5 −1 −0.5 0 0.5 1 1.5
−1
−0.5
0
0.5
1
Guinea pig
aardvark
alligator
alpaca
amphibiananemone
ant
anteater
antelope
ape
armadillo
baboon
badger
bat
bear
beaver
bee
beetle
bird
bison
bluejay
boaconstrictor
bobcat
buffalo
bull
bumblebee
bunny
butterfly
buzzard
camel
canary
cardinal
caribou
cat caterpillar
centipede
chameleon
cheetah
chickadeechicken
chimp
chinchilla
chipmunk
clam
cobra
cockatiel
cockroach
colt
cougarcow
coyote
crab
crane
cricket
crocodile
crow
deer
dingo
dog
dolphin
donkey
dove
duck
eagle
eel
elephant
elk
emu
falcon
fish
flamingo
flea fly
fowl
fox frog
gazelle
gecko
gerbilgibbon
gilamonster
giraffe
gnat
goat
goose
gopher
gorilla
grasshopper
groundhog
grouse
gull
hamsterhare
hawk
hedgehog
heron
hippopotamus
hornet
horse
human
hummingbird
hyena
ibis
iguana
impala insect
jackal
jackrabbit
jaguar
kangaroo
koalaladybug
lamb
lemur
leopard
lion
lizard
llama
lobster
loon
lynx
manatee
maremarten
mole
mongoose
monkey
moose
mosquito
moth
mouse
mule
muskox
muskrat
musky
newt
ocelot
octopus
opossum
orangutan
orioleostrich
otter
owl
oxpanda
panther
parrot
peacock
pelican
penguin
pig
pigeon
platypus
polar bear
pony
porcupine
porpoise
prairie dog
puma
python
quail
rabbitraccoon
racoon
ram
rat
rattlesnake
raven
reindeer
rhinoceros
roadrunner
robin
rodent
rooster
salamander
sea lion
seagull
seal
shark
sheepshrew
shrimp
skunk
sloth
snail
snake
sparrow
spider
squid
squirrel
swan
tadpole
tapir
tasmanian devil
tiger
tortoise
turkey
turtle
vole
vulture
wallaby
walrus
waspweasel
whale
wildebeest
wolf
wolverine
wombat
woodchuck
woodpecker
worm
wren
yak
zebra
K = 12, obj = 14.1208
5 / 30
![Page 6: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/6.jpg)
Memory search
−1.5 −1 −0.5 0 0.5 1 1.5
−1
−0.5
0
0.5
1
Guinea pig
aardvark
alligator
alpaca
amphibiananemone
ant
anteater
antelope
ape
armadillo
baboon
badger
bat
bear
beaver
bee
beetle
bird
bison
bluejay
boaconstrictor
bobcat
buffalo
bull
bumblebee
bunny
butterfly
buzzard
camel
canary
cardinal
caribou
cat caterpillar
centipede
chameleon
cheetah
chickadeechicken
chimp
chinchilla
chipmunk
clam
cobra
cockatiel
cockroach
colt
cougarcow
coyote
crab
crane
cricket
crocodile
crow
deer
dingo
dog
dolphin
donkey
dove
duck
eagle
eel
elephant
elk
emu
falcon
fish
flamingo
flea fly
fowl
fox frog
gazelle
gecko
gerbilgibbon
gilamonster
giraffe
gnat
goat
goose
gopher
gorilla
grasshopper
groundhog
grouse
gull
hamsterhare
hawk
hedgehog
heron
hippopotamus
hornet
horse
human
hummingbird
hyena
ibis
iguana
impala insect
jackal
jackrabbit
jaguar
kangaroo
koalaladybug
lamb
lemur
leopard
lion
lizard
llama
lobster
loon
lynx
manatee
maremarten
mole
mongoose
monkey
moose
mosquito
moth
mouse
mule
muskox
muskrat
musky
newt
ocelot
octopus
opossum
orangutan
orioleostrich
otter
owl
oxpanda
panther
parrot
peacock
pelican
penguin
pig
pigeon
platypus
polar bear
pony
porcupine
porpoise
prairie dog
puma
python
quail
rabbitraccoon
racoon
ram
rat
rattlesnake
raven
reindeer
rhinoceros
roadrunner
robin
rodent
rooster
salamander
sea lion
seagull
seal
shark
sheepshrew
shrimp
skunk
sloth
snail
snake
sparrow
spider
squid
squirrel
swan
tadpole
tapir
tasmanian devil
tiger
tortoise
turkey
turtle
vole
vulture
wallaby
walrus
waspweasel
whale
wildebeest
wolf
wolverine
wombat
woodchuck
woodpecker
worm
wren
yak
zebra
K = 12, obj = 14.1208
6 / 30
![Page 7: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/7.jpg)
Censored random walk
[Abbott, Austerweil, Griffiths 2012]
I V: n animal word types in English
I P : (dense) n× n transition matrix
I Censored random walk: observing only the first token of eachtype x1, x2, . . . , xt, . . .⇒ a1, . . . , an
I (star example)
7 / 30
![Page 8: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/8.jpg)
The estimation problem
Given m censored random walksD =
{(a(1)1 , ..., a
(1)n
), ...,
(a(m)1 , ..., a
(m)n
)}, estimate P .
8 / 30
![Page 9: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/9.jpg)
Each observed step is an absorbing random walk
(with Kwang-Sung Jun)
I P (ak+1 | a1, . . . , ak) may contain infinite latent steps
I Instead, model this observed step as an absorbing randomwalk with absorbing states V \ {a1, . . . , ak}
I P ⇒(Q R0 I
)
9 / 30
![Page 10: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/10.jpg)
Maximum Likelihood
I Fundamental matrix N = (I−Q)−1, Nij is the expectednumber of visits to j before absorption when starting at i
I p(ak+1 | a1, . . . , ak) =∑k
i=1NkiRi1
I log likelihood∑m
i=1
∑nk=1 log p(a
(i)k+1 | a
(i)1 , ..., a
(i)k )
I Nonconvex, gradient method
10 / 30
![Page 11: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/11.jpg)
Other estimators
I PRW: pretend a1, . . . , an not censored
I PFirst2: Use only a1, a2 in each walk (consistent)
11 / 30
![Page 12: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/12.jpg)
Star graphx-axis: m, y-axis: ‖P − P‖2F
12 / 30
![Page 13: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/13.jpg)
2D Grid
13 / 30
![Page 14: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/14.jpg)
Erdos-Renyi with p = log(n)/n
14 / 30
![Page 15: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/15.jpg)
Ring graph
15 / 30
![Page 16: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/16.jpg)
Questions
I Consistency?
I Rate?
16 / 30
![Page 17: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/17.jpg)
Outline
Human Memory Search
Machine Teaching
17 / 30
![Page 18: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/18.jpg)
Learning a noiseless 1D threshold classifier
θ∗
x ∼ uniform[0, 1]
y =
{−1, x < θ∗
1, x ≥ θ∗
Θ = {θ : 0 ≤ θ ≤ 1}
Passive learning:
1. given training data D = (x1, y1) . . . (xn, yn)iid∼ p(x, y)
2. finds θ consistent with D
Risk |θ − θ∗| = O(n−1)
18 / 30
![Page 19: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/19.jpg)
Learning a noiseless 1D threshold classifier
θ∗
x ∼ uniform[0, 1]
y =
{−1, x < θ∗
1, x ≥ θ∗
Θ = {θ : 0 ≤ θ ≤ 1}
Passive learning:
1. given training data D = (x1, y1) . . . (xn, yn)iid∼ p(x, y)
2. finds θ consistent with D
Risk |θ − θ∗| = O(n−1)
18 / 30
![Page 20: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/20.jpg)
Learning a noiseless 1D threshold classifier
θ∗
x ∼ uniform[0, 1]
y =
{−1, x < θ∗
1, x ≥ θ∗
Θ = {θ : 0 ≤ θ ≤ 1}
Passive learning:
1. given training data D = (x1, y1) . . . (xn, yn)iid∼ p(x, y)
2. finds θ consistent with D
Risk |θ − θ∗| = O(n−1)
18 / 30
![Page 21: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/21.jpg)
Active learning (sequential experimental design)
Binary searchθ∗
Risk |θ − θ∗| = O(2−n)
19 / 30
![Page 22: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/22.jpg)
Active learning (sequential experimental design)
Binary searchθ∗
Risk |θ − θ∗| = O(2−n)
19 / 30
![Page 23: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/23.jpg)
Machine teaching
What is the minimum training set a helpful teacher can construct?
θ∗
Risk |θ − θ∗| = ε,∀ε > 0
20 / 30
![Page 24: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/24.jpg)
Machine teaching
What is the minimum training set a helpful teacher can construct?θ∗
Risk |θ − θ∗| = ε,∀ε > 0
20 / 30
![Page 25: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/25.jpg)
Comparing the three
θ∗
O(1/2 )n
θ∗θ∗
{{
O(1/n)
passive learning "waits" active learning "explores" teaching "guides"
The teacher knows θ∗ and the learning algorithm.
21 / 30
![Page 26: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/26.jpg)
Example 2: Teaching a Gaussian distribution
Given a training set x1 . . . xn ∈ Rd, let the learning algorithm be
µ =1
n
n∑i=1
xi
Σ =1
n− 1
n∑i=1
(xi − µ)(xi − µ)>
How to teach N(µ∗,Σ∗) to the learner quickly?
non-iid, n = d+ 1
22 / 30
![Page 27: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/27.jpg)
Example 2: Teaching a Gaussian distribution
Given a training set x1 . . . xn ∈ Rd, let the learning algorithm be
µ =1
n
n∑i=1
xi
Σ =1
n− 1
n∑i=1
(xi − µ)(xi − µ)>
How to teach N(µ∗,Σ∗) to the learner quickly?
non-iid, n = d+ 1
22 / 30
![Page 28: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/28.jpg)
An Optimization View of Machine Teaching
A(D)
θ∗
Α (θ )−1 ∗
A
Α−1
D
Θ
DD
minD∈D
|D|
s.t. A(D) = θ∗
The objective is the teaching dimension [Goldman, Kearns 1995] ofθ∗ with respect to A,Θ.
23 / 30
![Page 29: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/29.jpg)
Example 3: Works on Humans
Human categorization on 1D stimuli [Patil, Z, Kopec, Love 2014]
human training set human test accuracy
machine teaching 72.5%iid 69.8%
(statistically significant)
24 / 30
![Page 30: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/30.jpg)
New task: teaching humans how to label a graph
Given:
I a graph G = (V,E)
I target labels y∗ : V 7→ {−1, 1}I a label-completion cognitive model A (graph diffusion
algorithm) such as:I mincutI harmonic function [Z, Gharahmani, Lafferty 2003]I local global consistency [Zhou et al. 2004]I . . .
Find the smallest seed set:
minS⊆V
|S|
s.t. A(y∗(S)) = y∗
(inverse problem of semi-supervised learning)
25 / 30
![Page 31: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/31.jpg)
Example: A = local-global consistency [Zhou et al. 2004]
F = (1− α)(I − αD−1/2WD−1/2)−1y∗(S)
y = sgn
(F
(1−1
))
26 / 30
![Page 32: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/32.jpg)
|A−1(y∗)| is largeTurns out 4649 out of 216 = 65536 training sets S lead to thetarget label completion
27 / 30
![Page 33: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/33.jpg)
Optimal seed set 1: |S| = 3
28 / 30
![Page 34: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/34.jpg)
Optimal seed set 2: |S| = 3
29 / 30
![Page 35: Some applications in human behavior modeling (and open questions)](https://reader035.vdocuments.site/reader035/viewer/2022062600/586782ef1a28abcd408bce0d/html5/thumbnails/35.jpg)
Questions
I How does |S| relate to (spectral) properties of G?
I How to solve for S?
30 / 30