ML-NYC Speaker Series: Bin Yu

The Machine Learning New York City (ML-NYC) Speaker Series and Happy Hour is a monthly event for machine learning practitioners, researchers, and students to meet and watch talks from leading researchers in the field. Each event will feature a New York City-based speaker presenting their work, followed by a reception. The ML-NYC Speaker Series is open to anyone interested in machine learning research, and we encourage everyone to attend, whether you are just getting started in research or an expert in the field.

We are excited to welcome Bin Yu as our next speaker on Monday, September 23. The talk will begin at 4pm at the Flatiron Institute, followed by a reception. You must register on Eventbrite to attend the talk and reception.

Title: Veridical Data Science and PCS Uncertainty Quantification

Abstract: Data Science is central to AI and has driven most of the recent advances in science. Human judgment calls are ubiquitous at every step of the data science life cycle (DSLC): problem formulation, data cleaning, EDA, modeling, and reporting. Such judgment calls are often responsible for the “dangers” of AI by creating a universe of hidden uncertainties well beyond sample-to-sample uncertainty. To mitigate these dangers, veridical (truthful) data science is introduced based on three principles: Predictability, Computability and Stability (PCS). The PCS framework and documentation unify, streamline, and expand on the ideas and best practices of statistics and machine learning for reproducible science. In every step of the DSLC, PCS emphasizes reality checks through predictability, considers computability up front, and takes into account expanded uncertainty sources under stability including those from data curation/cleaning and algorithm choice to build more trust in data results. PCS will be showcased through collaborative research in cancer detection and seeking genetic drivers of a heart disease. We will end PCS uncertainty quantification (PCS-UQ) that addresses two other prominent sources of uncertainty in the DSLC from reasonable choices practitioners make in data cleaning and modeling stages (in addition to uncertainty arising from data collection). We will also compare with conformal inference when appropriate.

Bio: Dr. Bin Yu is Chancellor’s Distinguished Professor and Class of 1936 Second Chair in Statistics, EECS, and Computational Biology at UC Berkeley. Her research focuses on the practice and theory of statistical machine learning, veridical data science, and solving interdisciplinary data problems in neuroscience, genomics, and precision medicine. She and her team have developed algorithms such as iterative random forests (iRF), stability-driven NMF, and adaptive wavelet distillation (AWD) from deep learning models. She is a member of the National Academy of Sciences and of the American Academy of Arts and Sciences. She was a Guggenheim Fellow, and holds an Honorary Doctorate from The University of Lausanne.

Organizers: The ML-NYC Speaker Series is organized by professors David Blei, Joan Bruna, and Lawrence Saul, Flatiron Fellows Neha Wadia and Brett Larsen, and PhD students Claudia Shi and Noah Amsel. Please email [email protected] if you have any questions. You can also follow us on Twitter @MLNYCSeries, join our mailing list, or check out our website.

Protocols:

- By entering the building each person implicitly attests that they do not have symptoms consistent with COVID and they are not knowingly COVID positive.

- Enter with a government issued ID (you will need to show your ID, not the QR code from Eventbrite) when arriving at the building.

- There is NO food or drink allowed in the auditorium.

- Please do not arrive earlier than 30 minutes before the event.

The Flatiron Institute is located at 162 Fifth Ave. The entrance is on 21st street between 5 & 6 Ave and the IDA auditorium is located on the 2nd floor.

Host or Publisher ML-NYC Speaker Series and Happy Hour