diversity aware recommendation for human collectives

19
Diversity-Aware Recommendaon for Human Collecves Pavlos Andreadis, Sofia Ceppi, Michael Rovatsos, Subramanian Ramamoorthy School of Informacs, University of Edinburgh Robust Autonomy and Decisions group CISA Agents group (FOCAS) (ICT-2011.9.10), as a Collaborave Project (generic), under the 7th Framework programme, Grant agreement n. 600854. The SmartSociety project is supported by the European Commission, in the area "FET Proacve: Fundamentals of Collecve Adapve Systems" ECAI, DIVERSITY Workshop Hague – August, 29 2016

Upload: smart-society-project

Post on 25-Jan-2017

75 views

Category:

Science


3 download

TRANSCRIPT

Page 1: Diversity aware recommendation for human collectives

Diversity-Aware Recommendation for Human Collectives

Pavlos Andreadis, Sofia Ceppi, Michael Rovatsos, Subramanian RamamoorthySchool of Informatics, University of Edinburgh

Robust Autonomy and Decisions group

CISA Agents group

(FOCAS) (ICT-2011.9.10), as a Collaborative Project (generic), under the 7th Framework programme, Grant agreement n. 600854.

The SmartSociety project is supported by the European Commission, in the area "FET Proactive: Fundamentals of Collective Adaptive Systems"

ECAI, DIVERSITY WorkshopHague – August, 29 2016

Page 2: Diversity aware recommendation for human collectives

2

Sharing Economy Applications

Requests Potential allocation

Page 3: Diversity aware recommendation for human collectives

3

Ridesharing Example

13:35

14:00

S

D

13:35

14:00

12:00

17:00

Requests Potential allocation

Arrival:

Arrival:

Page 4: Diversity aware recommendation for human collectives

4

Diversity-Aware Recommendation

Page 5: Diversity aware recommendation for human collectives

5

Diversity-Aware Recommendation

Page 6: Diversity aware recommendation for human collectives

6

Diversity-Aware Recommendation

Page 7: Diversity aware recommendation for human collectives

7

Problems to Address

Selecting set of solutions

Aiding user coordination

Page 8: Diversity aware recommendation for human collectives

8

Selecting Set of Solutions

Multiple criteria

Goal: Adaptive trade-of of system-level utility and fairness

system-level utility:

fairness:Social Welfare

number of Allocated Passengers

number of Drivers

Page 9: Diversity aware recommendation for human collectives

9

Aiding User Coordination Users select according to solution utility

How? Taxation

Taxation scheme depends on user selection behaviour

– Noiseless

– Constant noise

– Logit noise

modify

Goal: Sponsor a solution using minimal taxation

Page 10: Diversity aware recommendation for human collectives

10

Constraints: feasibility

Generating the Recommendation Set

MILPsystem

Constraints: MILPsystem +

MILPfirst

MILPothersk-1 x

Constraints: MILPfirst + + taxation constraint

Page 11: Diversity aware recommendation for human collectives

11

Experiment Design

Metrics: System utility; Fairness; Num Passengers allocated; Num Drivers w. passengers

Evaluations performed after user selections.

Num of users (10, 20); percentage of which drivers (20, 30, 40 %);

Utility threshold (50, 75, 100 %);

User selection model (constant, logit); For logit noise, probability (60, 80 %).

100 experiment instances per configuration.

VS VS

with rejection

Set Recommendation Benchmark Allocation Benchmark

Page 12: Diversity aware recommendation for human collectives

12

Set Recommendation Benchmark,Logit Noise

(no

n-)

We can outperform the benchmark in

terms of both system utility and fairness.

Page 13: Diversity aware recommendation for human collectives

13

Set Recommendation Benchmark,Logit Noise

(no

n-)

We can outperform the benchmark in

terms of both system utility and fairness.

Page 14: Diversity aware recommendation for human collectives

14

Set Recommendation Benchmark,Logit Noise

(no

n-)

We can outperform the benchmark in

terms of both system utility and fairness.

Page 15: Diversity aware recommendation for human collectives

15

Set Recommendation Benchmark,Logit Noise

(no

n-)

We can outperform the benchmark in

terms of both system utility and fairness.

Page 16: Diversity aware recommendation for human collectives

16

Allocation Benchmark,Logit Noise

We can allow users to have a choice at no

cost to the system or users.

with rejection(n

on

-)

Page 17: Diversity aware recommendation for human collectives

17

Allocation Benchmark,Logit Noise

with rejection(n

on

-)

We can allow users to have a choice at no

cost to the system or users.

Page 18: Diversity aware recommendation for human collectives

18

We presented a methodology for aiding the coordination of user collectives, in the absence of agent communication.

Ongoing work Expand uncertainty to consider beliefs over preferences;

Incorporate active learning procedures in the MILPs;

Examine robustness to varying degrees of incorrect assumptions.

Set Recommendation requires explicitly handling the uncertainty in user behaviour;

Our procedure can match the performance of a direct allocation (given rejection);

We can allow users to have a choice at no cost to the system;

We allow for adaptively trading-of system-level utility and fairness.

Conclusions

Page 19: Diversity aware recommendation for human collectives

19

Why Diversity-Aware?

We present users with options, and we are robust to innacurate

representations of their preferences. Further, we are able to learn from

their choices. We can achieve this at no cost to the users or the collective,

and we can adaptively trade-of between collective and user-specific

criteria.