The workshop will be held December 4 virtually. Times are listed in EST.
| Time | Title |
| 9:00-9:10 | Opening remarks |
| 9:10-9:30 | Uri Shalit: When and How can Machine Learning be Used for Treatment Recommendation Based on Patient Records |
| 9:30-9:50 | Shalmali Joshi: Towards Robust, Safe and Ethical ML for Health |
| 9:50-10:10 | Discussion with Uri Shalit and Shalmali Joshi |
| 10:10-10:20 | Lorena Qendro: Best thematic paper award - Early Exit Ensembles for Uncertainty Quantification |
| 10:20-10:25 | Q&A with Lorena Qendro |
| 10:25-10:40 | Paper spotlights A Tuan Truong: How Transferable are Self-supervised Features in Medical Image Classification Tasks? Milan Kuzmanovic: Deconfounding Temporal Autoencoder: Estimating Treatment Effects over Time Using Noisy Proxies Chao Pang: CEHR-BERT: Incorporating temporal information from structured EHR data to improve prediction tasks |
| 10:40-11:30 | Poster session A |
| 11:30-12:10 | Lunch |
| 12:10-12:30 | Rory Sayres: Case studies of real-world medical AI deployment: What can ML and HCI lenses teach us, and each other? |
| 12:30-12:50 | Xiao Liu: Evidence Standards for AI in Healthcare |
| 12:50-13:10 | Karandeep Singh: Bringing Machine Learning Models to the Bedside |
| 13:10-13:40 | Discussion with Rory Sayres, Xiao Liu, and Karandeep Singh |
| 13:40-14:40 | Round table discussions |
| 14:40-14:50 | Break |
| 14:50-15:10 | Aisha Walcott-Bryant: The Great Potential of AI to Transform Healthcare: A COVID-19 Story |
| 15:10-15:30 | Sendhil Mullainathan: Ground Truth in Health: Nightingale Open Science |
| 15:30-15:50 | Discussion with Aisha Walcott-Bryant and Sendhil Mullainathan |
| 15:50-16:00 | Tianwei Yin: Best paper award - End-to-End Sequential Sampling and Reconstruction for MRI |
| 16:00-16:05 | Q&A with Tianwei Yin |
| 16:05-16:20 | Paper spotlights B Seongsu Bae: Question Answering for Complex Electronic Health Records Database using Unified Encoder-Decoder Architecture Bryan Gopal: 3KG: Contrastive Learning of 12-Lead Electrocardiograms using Physiologically-Inspired Augmentations Neeraj Wagh: Domain-guided Self-supervision of EEG Data Improves Downstream Classification Performance and Generalizability |
| 16:20-17:10 | Poster session B |
| 17:10-17:20 | Break |
| 17:20-17:40 | Yoshua Bengio: Generative Active Learning to Explore Therapeutic Space |
| 17:40-18:00 | Discussion with Yoshua Bengio |
| 18:00-18:10 | Closing remarks |