CIS Seminar: “Generative multitask learning mitigates target-causing confounding”
Wu and Chen Auditorium (Room 101), Levine Hall 3330 Walnut Street, Philadelphia, PA, United StatesWe propose a simple and scalable approach to causal representation learning for multitask learning. Our approach requires minimal modification to existing ML systems, and improves robustness to prior probability shift. The improvement comes from mitigating unobserved confounders that cause the targets, but not the input. We refer to them as target-causing confounders. These confounders induce […]