![Page 1: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/1.jpg)
AI for Social Impact:Learning & Planning in the Data to Deployment Pipeline
MILIND TAMBE
Director Center for Research on Computation and Society
Harvard University
&
Director “AI for Social Good”
Google Research India
![Page 2: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/2.jpg)
Date: 7/17/2020 2
AI and Multiagent Systems Research for Social Impact
Public Safety
and Security
Conservation Public Health
![Page 3: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/3.jpg)
Viewing Social Problems as Multiagent Systems
Key research challenge across problem areas:
Optimize Our Limited Intervention Resources
when
Interacting with Other Agents
Date: 7/17/2020 3
![Page 4: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/4.jpg)
Optimizing Limited Intervention Resources
Date: 7/17/2020 4
Stackelberg
security
games
Green
security
games
Sampled
social
networks
Public Safety & Security Conservation
Public Health
![Page 5: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/5.jpg)
Google Research Bangalore Director, AI for Social Good
Date: 7/17/2020 5
Public Health
ConservationEducation
AI for
Social Good
workshop
![Page 6: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/6.jpg)
Field tests&
deployment
Prescriptivealgorithm
Multiagent ReasoningIntervention
Immersion
Data Collection
Date: 7/17/2020 6
Predictivemodel
Learning/Expert input
Three Common ThemesMultiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships
Game
TheorySocial
networks
![Page 7: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/7.jpg)
Date: 7/17/2020 7
Three Common ThemesMultiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships
Field tests&
deployment
Prescriptivealgorithm
Multiagent ReasoningIntervention
Immersion
Data Collection
Predictivemodel
Learning/Expert input
Field test & deployment: Social impact is a key objective
Lack of data is a norm: Must be part of project strategy
![Page 8: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/8.jpg)
Date: 7/17/2020 8
Three Common ThemesMultiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships
![Page 9: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/9.jpg)
Outline: Overview of Past 14 Years of Research
Public Safety & Security: Stackelberg Security Games (brief)
Date: 7/17/2020 9
Conservation/Wildlife Protection: Green Security Games
Public Health: Influence maximization & social networks
▪ AAMAS,AAAI,IJCAI
▪ Real world evaluation
▪ PhD students & postdocs
![Page 10: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/10.jpg)
ARMOR Airport Security: LAX(2007)
Game Theory direct use for security resource optimization?
Glasgow: June 30, 2007
Date: 7/17/2020 10
Erroll Southers LAX Airport, Los Angeles
![Page 11: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/11.jpg)
Terminal #1 Terminal #2
Terminal #1 4, -3 -1, 1
Terminal #2 -5, 5 2, -1
Adversary
Game Theory for Security Resource Optimization
Defender
Date: 7/17/2020 11
New Model: Stackelberg Security Games
![Page 12: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/12.jpg)
Terminal #1 Terminal #2
Terminal #1 4, -3 -1, 1
Terminal #2 -5, 5 2, -1
Adversary
Game Theory for Security Resource Optimization
Defender
Date: 7/17/2020 12
New Model: Stackelberg Security Games
![Page 13: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/13.jpg)
13
Terminal #1 Terminal #2
Terminal #1 4, -3 -1, 1
Terminal #2 -5, 5 2, -1
Adversary
Defender
Optimization: Not 100% security; increase cost/uncertainty to attackers
Stackelberg: Defender commits to randomized strategy, adversary responds
Date: 7/17/2020
Security game: Played on targets, payoffs based on calculated losses
New Model: Stackelberg Security Games
Game Theory for Security Resource Optimization
![Page 14: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/14.jpg)
ARMOR at LAX
Basic Security Game Operation [2007]
14Date: 7/17/2020
Kiekintveld Pita
Target #1 Target #2 Target #3
Defender #1 2, -1 -3, 4 -3, 4
Defender #2 -3, 3 3, -2 ….
Defender #3 …. …. ….
Pr (Canine patrol, 8 AM @Terminals 2,5,6) = 0.17
Pr (Canine patrol, 8 AM @ Terminals 3,5,7) = 0.33
……
Mixed Integer Program
Canine Team Schedule, July 28
Term 1 Term 2 Term 3 Term 4 Term 5 Term 6 Term 7 Term 8
8 AM Team1 Team3 Team5
9 AM Team1 Team2 Team4
… … … … … … … … …
![Page 15: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/15.jpg)
1.. =i
ixts
jq
ix
ijR
Xi Qj
max
1=Qj
jq
Maximize defender
expected utility
Defender mixed
strategy
Adversary response
Security Game MIP [2007]
Payoffs Estimated From Previous Research
MqxCa ji
Xi
ij )1()(0 −−
Adversary best
response
15
Target #1 Target #2 Target #3
Defender #1 2, -1 -3, 4 -3, 4
Defender #2 -3, 3 3, -2 ….
Defender #3 …. …. ….
Date: 7/17/2020
Kiekintveld Pita
i
j
![Page 16: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/16.jpg)
16
ARMOR:
Optimizing Security Resource Allocation [2007]
Date: 7/17/2020
January 2009•January 3rd Loaded 9/mm pistol•January 9th 16-handguns,
1000 rounds of ammo•January 10th Two unloaded shotguns •January 12th Loaded 22/cal rifle•January 17th Loaded 9/mm pistol•January 22nd Unloaded 9/mm pistol
First application: Computational game theory for operational security
![Page 17: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/17.jpg)
Massive Scale Security GamesLarge Number of Combinations: Guards to Targets
17Date: 7/17/2020
Attack
1
Attack
2
Attack
…
Attack
1000
1, 2, 3 .. 5,-10 4,-8 … -20,9
1, 2, 4 .. 5,-10 4,-8 … -20,9
1, 3, 5 .. 5,-10 -9,5 … -20,9
… 1041 rows
1000 targets, 20 guards: 1041 combinations
Kiekintveld Jain
![Page 18: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/18.jpg)
Deployed Security Games Systems…
Date: 7/17/2020 18
TSA testimony
Congressional subcommittee
US Coast Guard testimony
Congressional subcommittee
Erroll Southers testimony
Congressional subcommittee
2009 20112007
ARMOR IRIS PROTECT
![Page 19: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/19.jpg)
Reviewer 2 is not impressed!
Date: 7/17/2020 19
![Page 20: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/20.jpg)
Security Games superior in
Optimizing Limited Security Resources
Vs
Human Schedulers/“simple random”
Significant Real-World Evaluation Effort
Date: 7/17/2020 20
![Page 21: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/21.jpg)
▪ Game theory vs Baseline+Expert
Controlled
x
Not Controlled
0
5
10
15
20
# Captures /30min
# Warnings /30min
# Violations /30min
Game Theory
Baseline + Expert
0
25
50
75
100
Pre-ARMOR 2008 2009 2010
Miscellaneous Drugs Firearm Violations
Field Tests Against Adversaries
▪ 21 days of patrol, identical conditions
Date: 7/17/2020 21
Computational Game Theory in the Field
BeforeARMOR
![Page 22: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/22.jpg)
Outline
Public Safety & Security: Stackelberg Security Games
Date: 7/17/2020 22
Conservation/Wildlife Protection: Green Security Games
Public Health: Influence maximization/Game against nature
Dr Andy Plumptre
Conservation Biology
![Page 23: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/23.jpg)
Poaching of Wildlife in Uganda
Limited Intervention (Ranger) Resources to Protect Forests
23Date: 7/17/2020
Snare or Trap Wire snares
![Page 24: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/24.jpg)
Green Security Games[2015]
Limited Ranger Resources to Protect Forests
24Date: 7/17/2020
Fang
,max
. . 1
0 ( ) (1 )
x q ij i j
i X j Q
i
i
ij i j
i X
R x q
s t x
a C x q M
=
− −
Adversary not fully strategic; multiple “bounded rational” poachers
Defender
mixed strategy
Max defender
utility
![Page 25: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/25.jpg)
Green Security Games [2015]Game Theory + Machine Learning Poacher Behavior
25Date: 7/17/2020
Learn adversary bounded rational response: At each grid location i
Xu
Ranger patrols: X(i)
Features: F(i)
Probability
of finding
snare in
cell i
giMachine
Learning
max ( )
. . 1
xi X
ii
i ig x
s t x
=
Defender
mixed strategy
Max defender
utility
![Page 26: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/26.jpg)
Learning Adversary Model 12 Years of Past Poaching Data
Probability of snare
Per 1 KM Grid
Square
Ranger patrol
Animal density
Distance to rivers /
roads / villages
Area habitat
Area slope
…
26Date: 7/17/2020
Nguyen
gi
![Page 27: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/27.jpg)
Learning Adversary ModelUncertainty in Observations
27Date: 7/17/2020
Record: No Attack (NEG)
1 k
m
1 km
Record: Attack (POS)
1 km
1 k
mWalk more!
Nguyen
Probability of snare
Per 1 KM Grid
Square
Ranger patrol
Animal density
Distance to rivers /
roads / villages
Area habitat
Area slope
…
gi
![Page 28: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/28.jpg)
Date: 7/17/2020 28
Adversary Modeling [2016]
Imperfect Crime Observation-aware Ensemble Model
Patrol Effort
Predict: Ensemble of Classifiers
C0
0
500
1000
1500
PatrolEffort = 0
Tra
in D
ata NEG
POS
C1
0
500
1000
1500
PatrolEffort = 1
Tra
in D
ata
NEG
POS
C2
0
500
1000
1500
PatrolEffort = 2
Tra
in D
ata
NEG
POS
Training: Filtered Datasets
1 20
Gholami
![Page 29: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/29.jpg)
0
0.5
1
1.5
2
2.5
3
L&L Score
Train Labels SVM Bagging Ensemble Our Best Model
Poacher Behavior Prediction
PAWS: Protection Assistant for Wildlife Security
Poacher Attack Prediction in the Lab
Results from 2016
29Date: 7/17/2020
Gholami
![Page 30: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/30.jpg)
PAWS:Real-world Deployment 2016: First Trial
▪ Two 9-sq. km patrol areas
▪ Where there were infrequent patrols
▪ Where no previous hot spots
30Date: 7/17/2020
GholamiFord
![Page 31: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/31.jpg)
PAWS Real-world DeploymentTwo Hot Spots Predicted
▪ Poached Animals: Poached elephant
▪ Snaring: 1 elephant snare roll
▪ Snaring: 10 Antelope snares
Historical Base Hit
RateOur Hit Rate
Average: 0.73 3
31Date: 7/17/2020
GholamiFord
![Page 32: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/32.jpg)
PAWS Predicted High vs Low Risk Areas:2 National Parks, 24 areas each, 6 months [2017]
Date: 7/17/2020 32
H
L
0
0.2
0.4
0.6
Experiment Group
High-risk Medium-riskLow-risk
0
0.05
0.1
0.15
0.2
0.25
Experiment group
High-risk Low-risk
Queen Elizabeth National Park
Murchison Falls National Park
H
M
L
Snares per patrolled sq. KM Snares per patrolled sq. KM
Gholami
![Page 33: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/33.jpg)
PAWS Real-world Deployment Cambodia: Srepok Wildlife Sanctuary [2018-2019]
33Date: 7/17/2020
Xu
![Page 34: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/34.jpg)
PAWS Real-world Deployment Trials in Cambodia: Srepok National Park [2018-2019]
34Date: 7/17/2020
Xu
▪ 521 snares/month our tests
vs
▪ 101 snares/month 2018
0
0.1
0.2
0.3
0.4
Experiment Group
High-risk Medium-riskLow-risk
Snares per patrolled sq. KM
![Page 35: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/35.jpg)
Green Security Games:Around the Globe with SMART partnership [2019]
Date: 7/17/2020 35
Protect Wildlife800
National Parks Around the Globe
Also: Protect Forests, Fisheries…
![Page 36: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/36.jpg)
Green Security Games: Integrating Real-Time Information in the Pipeline
Data Collection
Date: 7/17/2020 36
Prediction
giPrescriptionmax ( )
. . 1
xi X
ii
i ig x
s t x
=
Field
Learn predictions with Historical Ground Truth Data
Real-Time information
![Page 37: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/37.jpg)
Green Security Games: Integrating Real-Time “SPOT” Information [2018]
Date: 7/17/2020 37
Goal: automatically find poachers
Bondi
![Page 38: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/38.jpg)
Drone Used to Inform Rangers [2019]
Prob(ranger) = 0.3
➢ Prob(ranger arrives) = 0.3 [poacher may not be stopped]
➢ Deceptive signaling to indicate ranger is arriving
BondiXu
![Page 39: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/39.jpg)
Drone Used to Inform Rangers [2019]
Prob(ranger) = 0.3
➢ Prob(ranger arrives) = 0.3 [poacher may not be stopped]
➢ Deceptive signaling to indicate ranger is arriving
BondiXu
![Page 40: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/40.jpg)
Drone Used to Inform Rangers [2019]
Prob(ranger) = 0.3
➢ Prob(ranger arrives) = 0.3 [poacher may not be stopped]
➢ Deceptive signaling to indicate ranger is arriving
➢ Must be strategic in deceptive signaling
BondiXu
![Page 41: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/41.jpg)
Strategic Signaling: Informational AdvantageDefender Knows Pure & Mixed Strategy
Xu
ranger
no ranger
0.30.3
0.70.4 0.4
0.6
No
Signal
New Model: Stackelberg Security Games with Optimal Deceptive Signaling
➢ Poacher best interest to “believe signal” even if know 50% time defender is lying
➢ Theorem: Signaling reduces complexity of equilibrium computation
![Page 42: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/42.jpg)
Strategic Signaling: Handling Detection ErrorExploit Informational Asymmetry to Mitigate Impact
Strategic signaling in presence of error in detecting adversaries
Bondi
➢ Signal selectively when no adversary detection
ranger
0.3
0.7
0.6
ranger
![Page 43: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/43.jpg)
➢ Not enough data
→ May not learn accurate enough adversary model
→ May lead to errors in planning patrols on targets
Green Security Games Pipeline:Not Enough Data in Some Parks
Date: 7/17/2020 43
➢ Game focused learning
→ Maximizing learning accuracy ≠ Maximizing decision quality
→ Learn to maximize decision quality
Perrault
![Page 44: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/44.jpg)
Minimize σ 𝑖∈𝑇 𝑞empirical log ො𝑞
Optimize
Previous Stage-by-Stage Method:Make Prediction as Accurate as Possible Then Plan
Perrault MateWilder
Maximize accuracy in Adversary target values estimates
Plan patrol Coverage
![Page 45: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/45.jpg)
Maximize accuracy in Adversary target values estimates
Plan patrol Coverage
Optimize
Stage by Stage Method:Need to Focus on Important Targets
Two targets:Large effect
Defender EU
Perrault MateWilder
Minimize σ 𝑖∈𝑇 𝑞empirical log ො𝑞
![Page 46: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/46.jpg)
Maximize accuracy only ofImportant targets
Plan patrol Coverage
Game-Focused Learning:Need to Focus on Important Targets
Two targets:Large effect
Defender EU
Perrault MateWilder
Minimize σ 𝑖∈𝑇 𝑞empirical log ො𝑞
![Page 47: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/47.jpg)
Perrault MateWilder
Maximize defender expected utility
Plan patrol Coverage
Optimize
Game-Focused Learning:End-to-End Method
Maximize defender’s expected utility
1− 𝑝𝑖 ො𝑞 𝑞empirical
![Page 48: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/48.jpg)
Previous Two-Stage Method:
Gradient Descent
Date: 7/17/2020 48
𝜕accuracy
𝜕weights=𝜕prediction
𝜕weights
𝜕accuracy
𝜕prediction➢ Max accuracy
gradient descent:
Perrault MateWilder
Prediction
gi
Data Collection
Model w/ weights
…
![Page 49: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/49.jpg)
Game-Focused Learning:
End-to-End Method
Date: 7/17/2020 49
➢ Game-focused
gradient descent:
𝜕obj(decision)
𝜕weights=𝜕prediction
𝜕weights
𝜕decision
𝜕prediction
𝜕obj(decision)
𝜕decision
Perrault MateWilder
Prediction
gi
Prescription
max ( )
. . 1
xi X
ii
i ig x
s t x
=
FieldData
Collection
Simulation
𝑓(𝑥)
Model w/ weights
![Page 50: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/50.jpg)
Date: 7/17/2020 50
➢ Focusing learning on important targets increases defender utility
Game-Focused Learning:End-to-End Method
![Page 51: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/51.jpg)
Outline
Public Safety & Security: Stackelberg Security Games
Date: 7/17/2020 51
Conservation/Wildlife Protection: Green Security Games
Public Health: Game against natureProf Eric Rice
Social Work
![Page 52: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/52.jpg)
Public HealthOptimizing Limited Intervention (Social Worker) Resources
52Date: 7/17/2020
Preventing HIV in homeless youth: Rates of HIV 10 times housed population
➢ Shelters: Limited number of peer leaders to spread HIV information in social networks
➢ “Real” social networks gathered from observations in the field; not facebook etc
![Page 53: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/53.jpg)
Influence Maximization Background
▪ Given:
▪ Social network Graph G
▪ Choose K “peer leader” nodes
▪ Objective:
▪ Maximize expected number of influenced nodes
▪ Assumption: Independent cascade model of information spread
53Date: 7/17/2020
![Page 54: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/54.jpg)
Independent Cascade Model and Real-world Physical Social Networks
54Date: 7/17/2020
A BP(A,B)=0.4
C D
0 1μ = 0.5
C D
0 1μ ∈ [0.3, 0.7]
![Page 55: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/55.jpg)
Robust, Dynamic Influence Maximization
▪ Worst case parameters: a zero-sum game against nature
▪ Payoffs: (performance of algorithm)/OPT
Nature
Chooses parameters
μ,σvs
Algorithm
Chooses policy, i.e.,
Chooses Peer leaders
55Date: 7/17/2020
Wilder
![Page 56: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/56.jpg)
Params #1 Params #2
Policy #1 0.8, -0.8 0.3, -0.3
Policy #2 0.7, -0.7 0.5, -0.5
HEALER Algorithm [2017]Robust, Dynamic Influence Maximization
▪ Equilibrium strategy despite exponential strategy spaces: Double oracle
Influencer’s oracle
Nature’s oracle
Params #1 Params #2 Params #3
Policy #1 0.8, -0.8 0.3, -0.3 0.4, -0.4
Policy #2 0.7, -0.7 0.5, -0.5 0.6, -0.6
Policy #3 0.6, -0.6 0.4, -0.4 0.7, -0.7
Influencer
Nature
56Date: 7/17/2020
Params #1 Params #2 Params #3
Policy #1 0.8, -0.8 0.3, -0.3 0.4, -0.4
Policy #2 0.7, -0.7 0.5, -0.5 0.6, -0.6
Policy #3 0.6, -0.6 0.4, -0.4 0.7, -0.7
Wilder
Theorem: Converge with approximation guarantees
\ Params #1 Params #2
Policy #1 0.8, -0.8 0.3, -0.3
Policy #2 0.7, -0.7 0.5, -0.5
Policy #3 0.6, -0.6 0.4, -0.4
![Page 57: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/57.jpg)
Challenge: Multi-step Policy
K = 4
1st time step
57Date: 7/17/2020
Params #1 Params #2 Params #3
Policy #1 0.8, -0.8 0.3, -0.3 0.4, -0.4
Policy #2 0.7, -0.7 0.5, -0.5 0.6, -0.6
Policy #3 0.6, -0.6 0.4, -0.4 0.7, -0.7
K = 4
2nd time step
WilderYadav
![Page 58: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/58.jpg)
7/17/2020 58
HEALER: POMDP Model for Multi-Step PolicyRobust, Dynamic Influence Maximization
Action
Choose nodes
Observation: Update
propagation probability
POMDP
Policy
HIDDEN STATE
Yadav
Params #1 Params #2 Params #3
Policy #1 0.8, -0.8 0.3, -0.3 0.4, -0.4
Policy #2 0.7, -0.7 0.5, -0.5 0.6, -0.6
Policy #3 0.6, -0.6 0.4, -0.4 0.7, -0.7
POMDP
partitions
![Page 59: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/59.jpg)
Pilot Tests with HEALER
with 170 Homeless Youth [2017]
Recruited youths:
HEALER HEALER++ DEGREE CENTRALITY
62 56 55
59Date: 7/17/2020
12 peer leaders
Yadav Wilder
![Page 60: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/60.jpg)
Results: Pilot Studies
7/17/2020 60
0
20
40
60
80
100
HEALER HEALER++ Degree
Percent of non-Peer Leaders
Informed Not Informed
![Page 61: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/61.jpg)
Data to Deployment Pipeline:Network Sampling to avoid Data Collection Bottleneck
61Date: 7/17/2020
Data collection costly Sample 18%
Sampling from largest
communities
Wilder
New experiment With 60 homeless youth
12 peer leaders
![Page 62: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/62.jpg)
Results: Pilot Studies
with New Sampling Algorithm [2018]
0
20
40
60
80
100
SAMPLE HEALER Degree
Percent of non-Peer Leaders
Informed Not Informed
62Date: 7/17/2020
Wilder
![Page 63: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/63.jpg)
Results of 900 Youth Study [RECENT]
7/17/2020 63
0
5
10
15
20
25
30
35
Perc
ent
red
uct
ion
Reduction in condomless sex
SAMPLE
DC
Control
![Page 64: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/64.jpg)
AI Assistant: HEALER
7/17/2020 64
![Page 65: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/65.jpg)
Next Steps: Data to Deployment PipelineUsing an RL agent?
65Date: 7/17/2020
State: Current
discovered graph RL agent
chooses node v
to query
Query budget
exceeded:
Run Influence
maximization
& get reward
Query budget
not
exceeded
![Page 66: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/66.jpg)
Data to Deployment Pipeline:Using an RL agent
66Date: 7/17/2020
State: Current
discovered graph RL agentchooses node v
to query
Budgetexceeded:
Influence
maximizationreward
Budgetnot
exceeded
Network Family
Improve %
Rural 23.76
Animal 26.6
Retweet 19.7
Homeless 7.91
![Page 67: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/67.jpg)
Public Health: Optimizing Limited Social Worker ResourcesPreventing Tuberculosis in India [2019]
67Date: 7/17/2020
Tuberculosis (TB): ~500,000 deaths/year, ~3M infected in India
➢ Non-adherence to TB Treatment
➢ Active case finding using social networks
![Page 68: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/68.jpg)
Non-Adherence to TB treatmentPreventing Tuberculosis in India [2019]
68Date: 7/17/2020
➢ Digital adherence tracking: Patients call phone #s on pill packs; many countries
➢ Health workers track patients on a dash board
➢ Predict adherence risk from phone call patterns? Intervene before patients miss dose
![Page 69: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/69.jpg)
TB Treatment Adherence but Limited Resources:
Intervening Selectively before patients miss doses
Date: 7/17/2020 69
Data Collect
Phone logs
Predicthigh risk patients
RF or LSTM
Prescription
ConstraintTop K
Field
Killian
➢ 15K patients, 1.5M calls
![Page 70: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/70.jpg)
Increasing TB Treatment Adherence:
Intervening before patients miss doses [2019]
Date: 7/17/2020 70
Killian
Data from
State ofMaharashtra
India
107
120
144
97
0
20
40
60
80
100
120
140
160
180
True Positives False Positives
Nu
mb
er o
f P
atie
nts
Best Model vs. Baseline: Prediction High Risk Patients
Baseline Best Model
+35% -19%
![Page 71: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/71.jpg)
Improving TB interventions
Decision-Focused vs Stage by Stage Methods
Date: 7/17/2020 71
Wilder
0.4
0.42
0.44
0.46
0.48
0.5
Interventions: Decision-Focused Better
Stage by Stage Decision Focused
Decision focused learning improves TB interventions
![Page 72: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/72.jpg)
Integrating with Everwell’s Platform
Date: 7/17/2020 72
Killian
This work has a lot of potential to save lives.
Bill ThiesCo-founder, Everwell Health Solutions
.
![Page 73: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/73.jpg)
Active Case Finding via Network Optimization
➢ Active case finding: Find key nodes in contact network to cure
➢ Active screening: How to allocate limited resourced?
➢ Work with Wadhwani AI
Ou
Total Infection
No intervention
Max-Degree
Max-Belief
Full-REMEDY
Active Case Finding in India
Perrault
![Page 74: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/74.jpg)
Suicide Prevention in Marginalized Populations: Choose Gatekeepers in social networks
7/17/2020 74
Nature
Chooses some
gatekeepers to not
participate
vs
Algorithm
Chooses K gatekeepers
▪ Worst case parameters: a zero-sum game against nature
Rahmattalabi
▪ Fairness of coverage
![Page 75: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/75.jpg)
Summary
AI & Multiagent Systems for Social Impact
Date: 7/17/2020 75
Cross-cutting challenge: How to optimize limited intervention resources
• Public safety & security, conservation, public health
Unifying themes
• Multiagent systems reasoning
• Data to deployment
Research contributions:
• Models, algorithms: Stackelberg Security Games, game-focused learning
• Beyond models and algorithms…
![Page 76: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/76.jpg)
Future: AI Research for Social Good
76Date: 7/17/2020
It is possible to simultaneously advance AI research & do social good
Important to step out of the lab and into the field
Embrace interdisciplinary research -- social work, conservation
Lack of data is the norm, a feature; part of the project strategy
AI for Social Impact should be evaluated differently
Data to deployment perspective: Not just improving algorithms
![Page 77: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/77.jpg)
Thank you!
77Date: 7/17/2020
Collaborate to realize AI’s tremendous potential to
Improving society & fighting social injustice
![Page 78: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/78.jpg)
Date: 7/17/2020 78
Game-Focused Learning:Reduces Errors on Important Targets
Game-focused Two-stage
Target Importance Target Importance
Erro
r
Erro
rEr
ror
![Page 79: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/79.jpg)
Field Evaluation of Schedule Quality
Train patrols: Game theory outperformed
expert humans schedule 90 officers
3
3.5
4
4.5
5
5.5
Q1
Q2
Q3
Q4
Q5
Q6
Q7
Q8
Q9
Q1
0
Q1
1
Q1
2
Securi
ty S
core
Human Game Theory
Date: 7/17/2020 79
Improved Patrol Unpredictability & Coverage for Less Effort
![Page 80: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/80.jpg)
Childhood Obesity Prevention
via Network Optimization
➢ Childhood obesity: Diabetes, stroke and heart disease
➢ Early intervention with mothers: Change diet/activity using social networks
➢ Competitive influences in networks: Add/remove edges for behavior change
Wilder Ou
![Page 81: AI for Social Impact - teamcore.seas.harvard.edu · Multiagent systems, Data-to-deployment pipeline, Interdisciplinary partnerships Field tests & deployment Prescriptive algorithm](https://reader033.vdocuments.us/reader033/viewer/2022042923/5f729fb9ab3ff2103b1171fc/html5/thumbnails/81.jpg)
Solving Problems: Overall Research FrameworkEnd-to-End, Data to Deployment Pipeline
Field tests&
deployment
Prescriptivealgorithm
Game theory
Intervention
Immersion
Data Collection
Date: 7/17/2020 81
Predictivemodel
Learning/Expert input