Machine Learning Summit
summary
Half-day of 30-minute lightning talks report concrete results: a cytometry pipeline using cytometry single-cell matrices and a forest of shape‑constrained trees plus Nextflow/Docker scales to millions of cells per patient and finds ~200–300 T‑cell populations, isolating two phenotypes that separate responders in a 27‑patient trial and validating on melanoma; social‑media classifiers (CNNs) hit ~95% F1 for informative tweets and use domain adaptation for new disasters; a fish‑farm vision stack processes ~1.5 petabytes/month to count sea lice, estimate weight, and detect feed waste at 99% accuracy; clinical NLP achieves AUC 0.82 and a number‑needed‑to‑screen ≈3.6 to flag surgical epilepsy candidates.
AI-written summary · about
chapters (9)
- 0:00Welcome and introduction
- 2:41Machine learning in oncology
- 31:17Disaster management and social media
- 35:39Industrial AI and IoT
- 56:21AI in sustainable aquaculture
- 1:22:22AI and deepfake threats
- 1:49:12AI-optimized indoor farming
- 2:15:43AI in pediatric research
- 2:40:52AI, ethics, and fairness
related talks
39mLightning Talks
27mTalk by Micah Sheller (Flower/ML Commons)
30mCultivating a company-wide machine learning culture
22mReimaging Healthcare
50mBuilding Intelligent Mobile Apps using Machine Learning
29mCancer diagnostics enabled by machines: finding and classifying single tumor....
01h 03mUsing Machine Learning and AI to serve people and communities
35mFrom cells to drug responses - machine learning in cancer research