DBTB INT Jill Desmond
summary
Works at VersaMe as a senior data scientist and describes an iterative workflow that abandons optimistic first passes and rebuilds data collection until labels are usable, arguing accurately labeled data is crucial because algorithms cannot learn from noisy ground truth. Describes an original project using machine learning for cochlear implants to detect and mitigate reverberation and a product that uses analytics to encourage parents to interact more with children, with lessons about painful but necessary relabeling cycles.
AI-written summary · about
related talks
21mDataEngBytes 2023 - SYD-T2-04 - Suneeta Mall
33mDataEngBytes 2023 - MEL-T1-02 - Kendra Vant
27mDataEngBytes 2023 - BRI-T1-07 - Joe Reis
39mDoing Data: The Critical Process of Data Preparation (...)
28mDataEngBytes 2023 - PER-T1-05 - Ellie Gholami
11mData-centric AI: A new paradigm
01h 03mUsing Machine Learning and AI to serve people and communities
49mdbt: A Software Engineering Approach To The Analytics Workflow
