van der Schaar Lab

University of Potsdam Kalman Lecture

Presentation

Mihaela van der Schaar will deliver the 4th annual Kalman Lecture at the University of Potsdam.

Title

Quantitative epistemology: conceiving a new human-machine partnership

Abstract

Quantitative epistemology is a new and transformational area of research pioneered by our lab as a strand of machine learning aimed at understanding, supporting, and improving human decision-making. We are developing machine learning models that  capture how humans acquire new information, how they pay attention to such information, how their beliefs may be represented, how their  internal models may be structured, how these different levels of knowledge are leveraged in the form of actions, and how such  knowledge is learned and updated over time. Because our approach is driven by observational data in studying knowledge as well as using machine learning methods for supporting and improving knowledge acquisition and its impact on decision-making, we call this “quantitative epistemology.

Our methods are aimed at studying human decision-making, identifying potential suboptimalities in beliefs and decision processes (such as cognitive biases, selective attention, imperfect retention of past experience, etc.), and understanding risk attitudes and their implications for learning and decision-making. This would allow us to construct decision support systems that provide humans with information pertinent to their intended actions, their possible alternatives and counterfactual outcomes, as well as other evidence to empower better decision-making.

Location and local date/time

This event will take place online on October 4 at 10:15 CEST (09:15 BST).

About the event

The Kalman Lecture is an annual public lecture and is named after Rudolf Emil Kálmán, the co-invented and developed of the Kalman Filter. The new annual lecture series will give insights into current scientific topics at the interface of machine learning, statistics and applied mathematics.

The event is finished.

Date

Oct 04 2021
Expired!

Time

Note: time is shown in BST
09:15 - 10:30
Category