BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//UM//UM*Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/Detroit
TZURL:http://tzurl.org/zoneinfo/America/Detroit
X-LIC-LOCATION:America/Detroit
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20070311T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20071104T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20261002T131533
DTSTART;TZID=America/Detroit:20261023T143000
DTEND;TZID=America/Detroit:20261023T163000
SUMMARY:Lecture / Discussion:Online Learning and Decision-Making for Dynamic Pricing and Performative Prediction
DESCRIPTION:Many modern decision systems are not passive observers of the world they measure: the decisions they make change the data they subsequently receive. For example\, a seller that posts a price influences the decisions of potential buyers\, and hence the sales data it goes on to col-\nlect\; a lender that deploys a credit-scoring rule changes the nature of the loan applications it receives. In such settings\, the classical separation\nbetween estimation and decision-making breaks down\, and statistical guarantees must be derived jointly with the policy that generates the\ndata. This dissertation develops estimation theory and online algorithms for two such feedback settings --- dynamic pricing and performative\nprediction --- with a unifying theme: shape-constrained and microfounded structure can replace the tuning parameters\, smoothness assump-\ntions\, and parametric models on which existing methods rely.\nThe first part concerns dynamic pricing. We first consider a monopolistic setting with censored demand\, in which the seller posts a price and\nobserves only whether a sale occurred. In the linear valuation model\, the customer’s valuation is linear in the features of the item offered\,\nup to market noise with unknown distribution F. Existing methods estimate F by kernel smoothing\, bandit techniques\, or upper confidence\nbounds\; all of them impose Lipschitz or stronger conditions on F\, and all of them leave the analyst to choose bandwidths\, confidence widths\,\nor regularization parameters. We instead exploit the fact that F is monotone and estimate it by isotonic regression. The resulting policy car-\nries no tuning parameters whatsoever\, assumes only that F is Hölder continuous\, and attains sublinear regret. It also outperforms existing\nmethods in simulations and on transaction data from Welltower Inc. We then move to a competitive setting in which N sellers post prices\nover T periods\, and each seller’s demand follows a single-index model with an unknown monotone\, s-concave link. We prove existence and\nuniqueness of the Nash equilibrium\, and give a semiparametric least-squares policy under which prices converge to the Nash equilibrium\,\nwhile each seller incurs sublinear regret.\nThe second part concerns performative prediction\, which involves a learner and a population of agents whose actions follow a distribution\nP. The learner deploys a model M that shifts this distribution from P to P(M)\, thereby changing the data the learner observes\; the map P(M)\nis called the distribution map. Much of the literature assumes that the distribution map is known\, whereas we learn it from agent-response\ndata. We study two settings: a discrete-action framework\, in which agents choose from a finite set of alternatives\, and a continuous-action\nframework\, in which an action may be any real number. For discrete actions\, we give a nonparametric estimator of the distribution map\ntogether with an exploration--exploitation scheme that we show to be effective in minimizing the performative risk. For continuous actions\,\nwe model agents’ responses as a cost-adjusted utility maximization problem and propose an estimator of the cost. Our approach leverages\noptimal transport to align the pre-deployment distribution P with the post-deployment distribution P(M). We provide a rate of convergence\nfor the proposed estimator and assess its quality through empirical demonstrations on a credit-scoring dataset.
UID:153213-21915429@events.umich.edu
URL:https://events.umich.edu/event/153213
CLASS:PUBLIC
STATUS:CONFIRMED
CATEGORIES:Dissertation
LOCATION:West Hall - 470
CONTACT:
END:VEVENT
END:VCALENDAR