Putting Machine Learning into Production: An Overview — Srijith Rajamohan, Databricks
Machine learning as a decision-making tool has gained wide acceptance. However, in order to extract information from data using machine learning, it is critical that the ML lifecycle be managed using best practices. In this talk, I will cover some of the foundational principles in scalable machine learning starting from data provenance and lineage and model building to model deployment and monitoring.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Dr. Srijith Rajamohan is a senior developer advocate at Databricks where he works on data science and machine learning problems. He is interested in all things related to the ML lifecycle and has a focus on natural language processing and Bayesian modeling. He recently launched a series of three courses on Coursera to teach computational statistics and Bayesian inference using PyMC3. He was a computational scientist from 2014 to 2020 at Virginia Tech, where he worked on natural language understanding using deep learning techniques, Bayesian inference and reproducible/scalable infrastructure for ML problems.
He is a graduate of the University of Tennessee SimCenter: Center of Excellence in Applied Computational Science and Engineering, where he did his thesis work on computational electromagnetics using the finite-element method and earned a doctorate in computational engineering in 2014. He also holds a master’s in electrical engineering from Pennsylvania State University, where he worked on implementing neural networks for computer vision on the IBM cell processor.
https://www.ischool.berkeley.edu/events/2022/putting-machine-learning-production-overview
Видео Putting Machine Learning into Production: An Overview — Srijith Rajamohan, Databricks канала Berkeley School of Information
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Dr. Srijith Rajamohan is a senior developer advocate at Databricks where he works on data science and machine learning problems. He is interested in all things related to the ML lifecycle and has a focus on natural language processing and Bayesian modeling. He recently launched a series of three courses on Coursera to teach computational statistics and Bayesian inference using PyMC3. He was a computational scientist from 2014 to 2020 at Virginia Tech, where he worked on natural language understanding using deep learning techniques, Bayesian inference and reproducible/scalable infrastructure for ML problems.
He is a graduate of the University of Tennessee SimCenter: Center of Excellence in Applied Computational Science and Engineering, where he did his thesis work on computational electromagnetics using the finite-element method and earned a doctorate in computational engineering in 2014. He also holds a master’s in electrical engineering from Pennsylvania State University, where he worked on implementing neural networks for computer vision on the IBM cell processor.
https://www.ischool.berkeley.edu/events/2022/putting-machine-learning-production-overview
Видео Putting Machine Learning into Production: An Overview — Srijith Rajamohan, Databricks канала Berkeley School of Information
Показать
Комментарии отсутствуют
Информация о видео
8 апреля 2022 г. 22:04:02
01:02:15
Другие видео канала
State of Data 2014: Data Science Teams in the Wild (DataEDGE 2014)I School Faculty Spotlight: Morgan AmesWhen Data Science Meets Design - Alan McConchie, Stamen Design (DataEDGE 2014)The I School in 2019: Where We’ve Been and Where We’re GoingRoundtable Discussion: Refusal of Surveillance Tech, Part 1 (April 12, 2021)Career Services: Networking TipsWhy your Big Data Initiative Sucks and What to do About it - DataEDGE 2015How to Scale AI-led Analytics — Umair Rauf (DataEDGE 2019)Toward Human-Centered Algorithmic Technologies (Min Kyung Lee)DataEDGE Conference: A new vision for data science — May 30--31, 2013WordSeer FeaturesInsight and Oversights: Shaping the Future of Visual Analytics with AI — Alvitta OttleySports Analytics and the Giants: Opportunities for Revenue Generation | DataEDGE 2016UC Berkeley School of Information Winter 2020 CommencementConstructing Experiments to Inform Business Innovation (DataEDGE 2014)Info 159/259. Natural Language ProcessingWomen in Data Science at UC Berkeley 2021: Data Science in ResearchTrainspotting and Predicting Train Delays | DataEDGE 2016Panel: Size Matters: Big Data, New Vistas in the Humanities and Social Sciences (DataEDGE 2012)At Scale and under Pressure: How Social Media Moderate, Choreograph, and Censor Public Discourse