Nathalie Baracaldo | Federated Learning
06/03/20 • 29 min
IBM Research recently introduced their perspective on a machine learning paradigm called Federated Learning in which multiple parties can all participate in training a single model with a shared goal. You can use data that is distributed between competitors, or even data distributed in one company across multiple geographies. They can participate in this so securely without sharing their raw data, and consequently get models that are much more generalizable than they would otherwise be able to achieve on their own.
Links related to this episode:
IBM Research recently introduced their perspective on a machine learning paradigm called Federated Learning in which multiple parties can all participate in training a single model with a shared goal. You can use data that is distributed between competitors, or even data distributed in one company across multiple geographies. They can participate in this so securely without sharing their raw data, and consequently get models that are much more generalizable than they would otherwise be able to achieve on their own.
Links related to this episode:
Previous Episode
Hector Dominguez | AI Ethics Series Part 2
Hector Dominguez PhD is the Open Data Coordinator for the city of Portland and part of the Smart City PDX team. Hector has led privacy and information initiatives for Portland focusing on use of ethical tools for technology solutions assessment, privacy and information protection principles. He has also worked on establishing Portland's citywide privacy and surveillance technology strategies/procedures as well as policy development for facial recognition and surveillance technologies.
Links Referenced
- Ethics Canvas Mapping: https://www.ethicscanvas.org/
- Social Responsibility: https://www.slideshare.net/HectorDominguez1/presentacion-social-responsibility
- Learn more about AI Fairness & Bias in this ebook: http://ibm.biz/BdqMvS
Next Episode
Cesar Mierzejek | Data Science Leadership Skills
Kinga and Cesar discuss his career at IBM and the skills he needed to learn to be a top data scientist. The key topics explored include: Learning to communicate the value of your work and why it matters, being succinct and focusing on what your stakeholder needs; although unintuitive, learning about product management as it has helped him focus on the customers for the model, and finally becoming a life long learner.
Links mentioned in this episode:
If you like this episode you’ll love
Episode Comments
Generate a badge
Get a badge for your website that links back to this episode
<a href="https://goodpods.com/podcasts/the-data-science-podcast-ibm-developer-podcasts-98896/nathalie-baracaldo-federated-learning-5229653"> <img src="https://storage.googleapis.com/goodpods-images-bucket/badges/generic-badge-1.svg" alt="listen to nathalie baracaldo | federated learning on goodpods" style="width: 225px" /> </a>
Copy