Wednesday, December 5, 2018 - 10:30am - 11:30am
Hamsa Bastani (Wharton School of the University of Pennsylvania)
Machine learning is increasingly used to inform consequential decisions. Yet, these predictive models have been found to exhibit unexpected defects when trained on real-world observational data, which are plagued with confounders and biases. Thus, it is critical to involve domain experts in an interactive process of developing predictive models; interpretability offers a promising way to facilitate this interaction.
Wednesday, October 3, 2018 - 2:45pm - 3:30pm
Velibor Misic (University of California, Los Angeles)
Optimal stopping is the problem of deciding when to stop a stochastic system to obtain the greatest reward; this arises in numerous application areas, such as finance, healthcare and marketing. State-of-the-art methods for high-dimensional optimal stopping involve determining an approximation to the value function or to the continuation value, and then using that approximation within a greedy policy.
Subscribe to RSS - interpretability