Learning from randomly arriving agents

Tho Ngoc Le; Vijay G. Subramanian; Randall A. Berry

doi:10.1109/ALLERTON.2017.8262737

Learning from randomly arriving agents

Tho Ngoc Le, Vijay G. Subramanian, Randall A. Berry

Electrical and Computer Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We add to a line of work considering the impact of observation imperfections in models of Bayesian observational learning. In particular, we study a discrete-Time model in which in each time-slot, an agent may randomly arrive. Agents who arrive have the opportunity to buy a given item. If an agent chooses to buy, this action is recorded for subsequent agents. However, the decisions of agents who do not choose to buy are not recorded. Hence, if no one buys in a given slot, agents are unaware if this was due to no agent arriving or an agent choosing not to buy. We study the impact of this uncertainty on the emergence of information cascades. Using a Markov chain based analysis, we show that incorrect cascades may occur and that the probability of such cascades is not monotonic in the arrival probability of a user. Moreover, if the agents' private signals are weak, wrong cascades are more likely to happen than correct cascades.

Original language	English (US)
Title of host publication	55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	196-197
Number of pages	2
ISBN (Electronic)	9781538632666
DOIs	https://doi.org/10.1109/ALLERTON.2017.8262737
State	Published - Jul 1 2017
Event	55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017 - Monticello, United States Duration: Oct 3 2017 → Oct 6 2017

Publication series

Name	55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017
Volume	2018-January

Other

Other	55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017
Country/Territory	United States
City	Monticello
Period	10/3/17 → 10/6/17

Funding

This work was supported by the NSF under grant IIS-1538827.

ASJC Scopus subject areas

Computer Networks and Communications
Hardware and Architecture
Signal Processing
Energy Engineering and Power Technology
Control and Optimization

Access to Document

10.1109/ALLERTON.2017.8262737

Cite this

Le, T. N., Subramanian, V. G., & Berry, R. A. (2017). Learning from randomly arriving agents. In 55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017 (pp. 196-197). (55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017; Vol. 2018-January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ALLERTON.2017.8262737

@inproceedings{035f8cffefa04b9e9884520d3780d158,

title = "Learning from randomly arriving agents",

abstract = "We add to a line of work considering the impact of observation imperfections in models of Bayesian observational learning. In particular, we study a discrete-Time model in which in each time-slot, an agent may randomly arrive. Agents who arrive have the opportunity to buy a given item. If an agent chooses to buy, this action is recorded for subsequent agents. However, the decisions of agents who do not choose to buy are not recorded. Hence, if no one buys in a given slot, agents are unaware if this was due to no agent arriving or an agent choosing not to buy. We study the impact of this uncertainty on the emergence of information cascades. Using a Markov chain based analysis, we show that incorrect cascades may occur and that the probability of such cascades is not monotonic in the arrival probability of a user. Moreover, if the agents' private signals are weak, wrong cascades are more likely to happen than correct cascades.",

author = "Le, {Tho Ngoc} and Subramanian, {Vijay G.} and Berry, {Randall A.}",

year = "2017",

month = jul,

day = "1",

doi = "10.1109/ALLERTON.2017.8262737",

language = "English (US)",

series = "55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "196--197",

booktitle = "55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017",

address = "United States",

note = "55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017 ; Conference date: 03-10-2017 Through 06-10-2017",

}

Le, TN, Subramanian, VG & Berry, RA 2017, Learning from randomly arriving agents. in 55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017. 55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017, vol. 2018-January, Institute of Electrical and Electronics Engineers Inc., pp. 196-197, 55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017, Monticello, United States, 10/3/17. https://doi.org/10.1109/ALLERTON.2017.8262737

Learning from randomly arriving agents. / Le, Tho Ngoc; Subramanian, Vijay G.; Berry, Randall A.
55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017. Institute of Electrical and Electronics Engineers Inc., 2017. p. 196-197 (55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017; Vol. 2018-January).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Learning from randomly arriving agents

AU - Le, Tho Ngoc

AU - Subramanian, Vijay G.

AU - Berry, Randall A.

PY - 2017/7/1

Y1 - 2017/7/1

N2 - We add to a line of work considering the impact of observation imperfections in models of Bayesian observational learning. In particular, we study a discrete-Time model in which in each time-slot, an agent may randomly arrive. Agents who arrive have the opportunity to buy a given item. If an agent chooses to buy, this action is recorded for subsequent agents. However, the decisions of agents who do not choose to buy are not recorded. Hence, if no one buys in a given slot, agents are unaware if this was due to no agent arriving or an agent choosing not to buy. We study the impact of this uncertainty on the emergence of information cascades. Using a Markov chain based analysis, we show that incorrect cascades may occur and that the probability of such cascades is not monotonic in the arrival probability of a user. Moreover, if the agents' private signals are weak, wrong cascades are more likely to happen than correct cascades.

AB - We add to a line of work considering the impact of observation imperfections in models of Bayesian observational learning. In particular, we study a discrete-Time model in which in each time-slot, an agent may randomly arrive. Agents who arrive have the opportunity to buy a given item. If an agent chooses to buy, this action is recorded for subsequent agents. However, the decisions of agents who do not choose to buy are not recorded. Hence, if no one buys in a given slot, agents are unaware if this was due to no agent arriving or an agent choosing not to buy. We study the impact of this uncertainty on the emergence of information cascades. Using a Markov chain based analysis, we show that incorrect cascades may occur and that the probability of such cascades is not monotonic in the arrival probability of a user. Moreover, if the agents' private signals are weak, wrong cascades are more likely to happen than correct cascades.

UR - http://www.scopus.com/inward/record.url?scp=85048141474&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048141474&partnerID=8YFLogxK

U2 - 10.1109/ALLERTON.2017.8262737

DO - 10.1109/ALLERTON.2017.8262737

M3 - Conference contribution

T3 - 55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017

SP - 196

EP - 197

BT - 55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 55th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2017

Y2 - 3 October 2017 through 6 October 2017

ER -

Learning from randomly arriving agents

Abstract

Publication series

Other

Funding

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this