Vector Researchers Co-Lead Inaugural ACM Conference on Health, Inference, and Learning
The inaugural ACM Conference on Health, Inference, and Learning (CHIL) launched virtually on July 23, 2020, co-led by Vector Institute researchers. Conceived by Vector Faculty Member Marzyeh Ghassemi, the conference targets interdisciplinary experts in machine learning, health policy, and clinical data. Originally planned for Toronto, the event featured keynotes from prominent figures like Yoshua Bengio and Ruslan Salakhutdinov. Vector Institute maintained significant leadership presence, with Anna Goldenberg on the steering committee and other staff handling logistics and communications. The conference emphasized innovative machine learning deployments in clinical settings, addressing challenges such as fairness, causality, and data sharing. Several Vector-affiliated researchers presented papers, including studies on quantifying biases in clinical word embeddings and improving data extraction pipelines for the MIMIC-III ICU dataset. These contributions highlight the institute's commitment to rigorous evaluation of model biases and reproducibility in health AI. The event served as a platform for discussing the transition of clinical machine learning from off-the-shelf adaptations to domain-specific innovations, fostering collaboration between academia and industry.
Wire timeline
Vector Researchers Co-Lead Inaugural ACM Conference on Health, Inference, and Learning
The inaugural ACM Conference on Health, Inference, and Learning (CHIL) launched virtually on July 23, 2020, co-led by Vector Institute researchers. Conceived by Vector Faculty Member Marzyeh Ghassemi, the conference targets interdisciplinary experts in machine learning, health policy, and clinical data. Originally planned for Toronto, the event featured keynotes from prominent figures like Yoshua Bengio and Ruslan Salakhutdinov. Vector Institute maintained significant leadership presence, with Anna Goldenberg on the steering committee and other staff handling logistics and communications. The conference emphasized innovative machine learning deployments in clinical settings, addressing challenges such as fairness, causality, and data sharing. Several Vector-affiliated researchers presented papers, including studies on quantifying biases in clinical word embeddings and improving data extraction pipelines for the MIMIC-III ICU dataset. These contributions highlight the institute's commitment to rigorous evaluation of model biases and reproducibility in health AI. The event served as a platform for discussing the transition of clinical machine learning from off-the-shelf adaptations to domain-specific innovations, fostering collaboration between academia and industry.
Vector Institute for Artificial Intelligence