Overabundant Information and Learning Traps

Liang, Annie; Mu, Xiaosheng

doi:10.2139/ssrn.3057805

Cited by 4 publications

(5 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Generically, the social planner will not be able to improve (at late periods) upon the information that has been aggregated so far. We generalize this qualitative insight in our companion piece Liang and Mu (2018) and demonstrate also how it can fail.…”

Section: Discussionsupporting

confidence: 52%

See 1 more Smart Citation

Optimal and Myopic Information Acquisition

Liang¹,

Mu²,

Syrgkanis³

2017

Preprint

Self Cite

View full text Add to dashboard Cite

We consider the problem of optimal dynamic information acquisition from many correlated information sources. Each period, the decision-maker jointly takes an action and allocates a fixed number of observations across the available sources. His payoff depends on the actions taken and on an unknown state. In the canonical setting of jointly normal information sources, we show that the optimal dynamic information acquisition rule proceeds myopically after finitely many periods. If signals are acquired in large blocks each period, then the optimal rule turns out to be myopic from period 1.These results demonstrate the possibility of robust and "simple" optimal information acquisition, and simplify the analysis of dynamic information acquisition in a widely used informational environment.

show abstract

Section: Discussionsupporting

confidence: 52%

“…Myopic information acquisition (from period 1) leads to exclusive sampling of signal X 1 , while a patient DM eventually samples only from X 2 and X 3 . This example is generalized in Liang and Mu (2018).…”

Section: Intuition For Theorems 1-3mentioning

confidence: 99%

Optimal and Myopic Information Acquisition

Liang¹,

Mu²,

Syrgkanis³

2017

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recently, the question of information acquisition in the presence of multiple information sources has been pursued among others by Che and Mierendorff (2016), Liang et al (2017), Liang and Mu (2018), Fudenberg et al (2017), and Mayskaya (2017). In contrast, in this paper we explore information acquisition from multiple sources of information in a principal-agent setting where the incentives of the two parties differ.…”

Section: Introductionmentioning

confidence: 99%

When and How to Reward Bad News

Kuvalekar

Ravi

2018

SSRN Journal

View full text Add to dashboard Cite

We examine when and how to reward the bearer of bad news in a dynamic principalagent relationship with experimentation. The agent receives flow rents from experimentation, and divides his time between searching for conclusive good news and conclusive bad news about project quality. The principal commits in advance to rewards conditional on the type of news. At each instant, the principal observes the agent's allocation and news and makes a firing decision. We show that the principal's optimal Markov perfect equilibrium features a stark reward structure: either the principal does not reward the bearer of bad news at all or rewards the bearer of either news equally.

show abstract

“…Several papers study "incentivizing exploration" in substantially different models: with a social network [6]; with time-discounted utilities [10]; with monetary incentives [16,19]; with a continuous information flow and a continuum of agents [15]; with long-lived agents and "exploration" separate from payoff generation [29,31,32]; with fairness [25]. Also, seminal papers [11,27] study scenarios with long-lived, exploring agents and no principal.…”

Section: Introductionmentioning

confidence: 99%

Bayesian Exploration with Heterogeneous Agents

Immorlica

Mao

Slivkins

et al. 2019

The World Wide Web Conference

View full text Add to dashboard Cite

It is common in recommendation systems that users both consume and produce information as they make strategic choices under uncertainty. While a social planner would balance "exploration" and "exploitation" using a multi-armed bandit algorithm, users' incentives may tilt this balance in favor of exploitation. We consider Bayesian Exploration: a simple model in which the recommendation system (the "principal") controls the information flow to the users (the "agents") and strives to incentivize exploration via information asymmetry. A single round of this model is a version of a well-known "Bayesian Persuasion game" from [24]. We allow heterogeneous users, relaxing a major assumption from prior work that users have the same preferences from one time step to another. The goal is now to learn the best personalized recommendations. One particular challenge is that it may be impossible to incentivize some of the user types to take some of the actions, no matter what the principal does or how much time she has. We consider several versions of the model, depending on whether and when the user types are reported to the principal, and design a near-optimal "recommendation policy" for each version. We also investigate how the model choice and the diversity of user types impact the set of actions that can possibly be "explored" by each type.Additional Key Words and Phrases: bayesian exploration, incentivizing exploration, heterogeneous agents ACM Reference Format:

show abstract

Overabundant Information and Learning Traps

Cited by 4 publications

References 36 publications

Optimal and Myopic Information Acquisition

Optimal and Myopic Information Acquisition

When and How to Reward Bad News

Bayesian Exploration with Heterogeneous Agents

Contact Info

Product

Resources

About